- Rating
- 3(+1)
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
The paper defines "expression" as the average absolute GELU activation of the MLP unit over 46 fixed sentences, defines "initial memory" as the correlation between the current expression of each unit and its initial expression, and characterizes the decay time by the number of steps where the correlation is less than 0.10. The authors establish a Pythia map covering five scales, ten training sessions for each scale, and four larger models. They then use a re-trained small model to conduct experiments on transplanting initial working points, adjusting scale/learning rate, and resetting units. The main observation is that the projection of the early expression onto the average input across layers is closely related to the input weights; as the average input direction rotates, the initial correlation of the expression first decays from the deeper layers, while the projection of the weights in the original direction remains preserved; rescaling the cumulative learning rate can better align the controlled memory curve. After training with 14M damage, the training loss and the set of defined expression trajectories are close to recovery, but the unit identities change; the expression trajectory set of 70M does not recover in the same way.
Critical strengths
-
Decompose a broad initialization phenomenon into directly verifiable unit-level relationships. At step 128 of 14M, the current weights still have an initial projection correlation of .99 when read against the initial mean, while the initial weights drop to .49 when read against the current mean, approaching an expression memory of .38; the hierarchical expression memories of four scales are approximately .98 in correlation with the cosine of the mean direction, and the inter-layer correlations within the same step are also very high. The decomposition, random projection identity, and actual expression readings are interconnected, providing a more detailed description than "small weight changes". The paper also clearly distinguishes the arithmetic verification of the standard random projection identity from the empirical findings for nonlinear expressions, and does not directly attribute the recovery of the constrained readings to the causal proof of the training dynamics.
-
Self-checking of the instrument and comparison baseline substantially improves the credibility of the conclusion. The maps of five scales with ten seeds each are complementary to the initial value twin, orthogonal perturbation, repeated execution, different initial values, and intra-layer neighbor baselines. Especially in Appendix D: Removing step 0 still leaves almost the same early checkpoints as it was; after disrupting the later unit identities, the fate classification still has the explanatory power for the initial expression. This is a specific and operational failure diagnosis. The author also reports the stability of half sentences and all sentences, and acknowledges that this cannot replace the independent corpus verification; these controls make the applicable boundaries of the observational results clearer.
Critical weaknesses
The curve collapse supports an approximate scaling relationship, but is not sufficient to support the claim that "the learning rate and initial scale only act through this ratio". The controlled experiments only changed the multiplier of the same learning rate schedule, and 70M all stopped at the end of the same linear warmup. At this point, the cumulative learning rate was approximately proportional to "multiplier × number of steps squared", so these experiments can support the time rescaling of the tested schedule family, but cannot distinguish whether different schedule histories are equivalent when the cumulative value is the same. On the other hand, Appendix B.8 has reported that the minimum value is relatively flat, and the range of 14M contains k=1. Fixing k=1 at 14M/70M still reduces the divergence to 18.4%/17.4% of the original value, which is close to the optimal 17%/13%. Using 1.2/1.4 as an exact new index requires caution; currently, there is a lack of a clear interpolation grid for the collapsed objective function, the common intervals of each condition, and the weighting method. This doubt does not deny the controlled observation of the delayed decay of the initial scale increase, but limits "the sole clock" and the intensity of the index increment.
What remains unchanged/recovered is the measured loss and expression statistics, but it cannot be concluded that the predictive function remains unchanged or that the general functional role has been restored. §3 infers that "the content calculated by the network" has not changed based on the training loss change not exceeding 0.008 nats after transplantation; however, Appendix B.3 moves the input working point without corresponding replacement of output weights, which is not maintaining the function's neuron re-labeling. Similarly, §5.2's Hungarian matching uses a single average expression trajectory across training steps, discarding the responses to which sentences, tokens, and their symbols; a low matching distance can explain that this trajectory set is close to twin, but is not sufficient to prove that the same function is replaced by different units. The authors have limited 70M, the number of seeds, and the time window, and these limitations should be retained; the "recovery of calculation/role redistribution" should be clearly constrained to the defined loss and expression trajectory results.
Use the existing curves for limited validation. Please fully define the curve collapse loss, interpolation, common U_k interval, stage weights, and bootstrap objects, and compare the results of k=1 and the fitted k under a fixed common interval; can the fitted index be evaluated using the leave-one initialization method to improve the curve not involved in the fitting? If the improvement is still stable, especially at 70M, it can strengthen the index conclusion; if the main benefit comes from changing the comparison interval or fitting degrees of freedom, the result should be positioned as an approximate collapse compatible with k≈1. Do you agree that "only through the cumulative learning rate" is applicable to the current isomorphic schedule family?
The measurement objects of the conclusion. "Not changing the content calculated by the network" and "role redistribution" are operational names for the training loss and the expression trajectory represented by Equation 1, or do they contain a stronger assertion of the response to each input-output/function identity? If the former, the summary-style generalization, §3, and §5.2's wording can be tightened uniformly; if the cache can support the latter, provide corresponding output differences or sentence-by-sentence response results, and compare them with the existing twin/replicated run baseline. This answer will change the credibility of the evidence I give for the functional explanation, but does not require the addition of large-scale task benchmarks.
Questions for authors
see weeknesses
7 comments
Where the memory falls to 0.10 (initializations 0 / 1 / 2)
| Schedule | Learning rate at the start | 14M: step | 14M: accumulated learning rate | 70M: step | 70M: accumulated learning rate |
|---|---|---|---|---|---|
| Warmup over 4,000 steps (submitted) | rises from zero | 2,772 / 2,925 / 4,922 | 961 / 1,069 / 2,888 | 1,406 / 1,584 / 1,458 | 247 / 314 / 266 |
| Warmup over 1,000 steps, then constant | rises from zero | 1,376 / 1,542 / 2,977 | 876 / 1,042 / 2,477 | 696 / 837 / 623 | 242 / 351 / 194 |
| Quarter rate for 2,000 steps, then peak | a quarter of the peak | 2,043 / 2,306 / 3,899 | 543 / 806 / 2,399 | 752 / 1,387 / 922 | 188 / 347 / 231 |
| Constant | the peak | 122 / 150 / 714 | 122 / 150 / 714 | 15 / 14 / 15 | 15 / 14 / 15 |
| Peak for 500 steps, then a tenth | the peak | 122 / 150 / 2,151 | 122 / 150 / 665 | 15 / 14 / 15 | 15 / 14 / 15 |
Early in the decay the schedules agree: at accumulated learning rates of 2, 8 and 32 the five 14M curves are within 0.05 of each other, at steps up to 63 times apart. Later they separate according to the rate at which they start. The two warmup schedules stay within 0.03 of each other at every value of the first table, and a schedule that starts at the peak rate reaches a memory of 0.10 at an accumulated learning rate 4 to 8 times smaller than the submitted warmup. The accumulated learning rate thus removes a large part of the difference between schedules (the largest gap between two schedules is a factor of 7 to 23 in steps and a factor of 4 to 8 in accumulated learning rate), and not all of it.
At 70M the 0.10 crossing gives the same picture, and the separation is larger. The two warmup schedules reach a memory of 0.10 at accumulated learning rates within a factor of 1.4 of each other, at steps a factor of 1.9 to 2.3 apart. A schedule that starts at the peak rate is already below 0.10 at step 16; it reaches 0.10 at an accumulated learning rate 17 to 22 times smaller than the submitted warmup (94 to 113 times in steps).
So the answer to the reviewer's question is yes. The revision narrows the claim to the family of warmup schedules, where the collapse is approximate. This does not affect the conclusions of the paper. The paper describes the early phase that warmup produces (Section 7), and every Pythia run and every run of the paper uses a warmup. Within warmup schedules the new runs support the collapse at both sizes, for two warmup lengths. The other three findings do not depend on the collapse.
W1.3 "… there is a lack of a clear interpolation grid for the collapsed objective function, the common intervals of each condition, and the weighting method."
W3.2 "Please fully define the curve collapse loss, interpolation, common U_k interval, stage weights, and bootstrap objects …"
For one initialization there are five conditions (unmodified; block weights scaled by ; learning rate scaled by ). Every logged step of condition is placed at
where is the schedule's multiplier of the peak learning rate at step ; this is . Each memory curve is linearly interpolated in onto 50 points spaced uniformly over the interval on which all five curves are defined, . The spread is the variance across the five conditions at each grid point, averaged over the grid and then over initializations. Stages therefore carry no weights of their own: the grid weighs uniformly. The "raw" spread is the same quantity with . The reported exponent minimizes over . The bootstrap resamples initializations with replacement (1,000 draws) and refits; with three initializations it is essentially their range, as the appendix says. In the submission this interval depends on . The comparison below fixes it, using the intersection of these intervals over every of the grid.
W1.4 "Fixing k=1 at 14M/70M still reduces the divergence to 18.4%/17.4% of the original value, which is close to the optimal 17%/13%."
W3.3 "… compare the results of k=1 and the fitted k under a fixed common interval …"
against the best
| Objective | Quantity | 14M | 70M |
|---|---|---|---|
| Submitted | Best | 1.20 | 1.40 |
| Spread at the best , relative to the step axis | 0.17 | 0.13 | |
| Spread at , relative to the step axis | 0.18 | 0.17 | |
| Interval fixed | Best | 1.20 | 1.40 |
| Spread at , relative to the best | 1.10 | 1.43 |
14M has three initializations and 70M has five. With one interval fixed for every , the best does not change, and gives 1.10 times the smallest spread at 14M and 1.43 times at 70M. At 70M the gain over therefore does not come from the moving interval.
W3.4 "… can the fitted index be evaluated using the leave-one initialization method to improve the curve not involved in the fitting?"
Yes. We fitted on all initializations but one (submitted objective) and scored the one left out. The fitted does better than on the held-out curve in two of three cases at 14M and in all five at 70M. Relative to , the held-out spread is 0.85 to 1.44 at 14M and 0.72 to 0.81 at 70M.
W1.5 "Using 1.2/1.4 as an exact new index requires caution …"
W3.5 "If the improvement is still stable, especially at 70M, it can strengthen the index conclusion; if the main benefit comes from changing the comparison interval or fitting degrees of freedom, the result should be positioned as an approximate collapse compatible with k≈1."
The dependence on the initial scale is stronger than linear at 70M and compatible with linear at 14M. The reviewer's test separates the two sizes: at 70M the improvement is stable under both checks, and at 14M it is not. The numerical value of the exponent depends on how the collapse is scored. A statistic that needs no interval, the step at which each curve passes 0.8, 0.6, 0.4, 0.3 and 0.2 (with from 0 to 3), gives a best of 1.8 at 14M and 2.0 at 70M, against 1.2 and 1.4 by the spread. The revision therefore reports the direction at 70M and does not present 1.2 and 1.4 as measured constants. The exponent is a detail of the third finding: the abstract does not state it, and the other three findings do not depend on it.
W1.6 "This doubt does not deny the controlled observation of the delayed decay of the initial scale increase, but limits 'the sole clock' and the intensity of the index increment."
We share this reading. The controlled observation stands, and the revision narrows two statements. The abstract and Section 4.2 will state the result as an approximate collapse on accumulated learning rate over initial scale within the family of warmup schedules, which replaces "the learning rate enters only through its sum" in Section 4.2 and the matching row of Table 1. Section 4.2 will describe the dependence on the initial scale as compatible with at 14M and stronger at 70M, which replaces the measured exponent where the paper lists it as a result (Section 6, Section 7, Appendix A and Table 1). Appendix B.8 will carry the full definition above, the fixed-interval and held-out comparisons, the interval-free statistic, and the schedule experiment. What a unit inherits and how the inherited state is lost rest on neither statement.
W2 and W4. What is restored, the loss or the function
W2.1 "What remains unchanged/recovered is the measured loss and expression statistics, but it cannot be concluded that the predictive function remains unchanged or that the general functional role has been restored. …"
The reviewer is right about what the two experiments measure. The measurements in the submission stand: after the transplant the loss is unchanged, and after the damage the loss recovers and the repertoire of mean-expression trajectories returns. What was not aligned was the wording, since the two phrases that the reviewer quotes read as statements about the function, while the measurements concern the loss and Equation 1. To resolve this we reran both experiments with the weights saved and read the predictions and the responses of single units. The revision states each result at the level that was measured, and the contribution of the paper is not affected.
W4.1 "'Not changing the content calculated by the network' and 'role redistribution' are operational names for the training loss and the expression trajectory represented by Equation 1, or do they contain a stronger assertion of the response to each input-output/function identity?"
In the submission they were operational names for the training loss and for the mean-expression trajectories of Equation 1.
We thank the reviewer for a precise review. We ran the proposed checks on the existing curves, and new experiments where the existing curves could not decide. The four paragraphs under Critical Weaknesses (W1 to W4, in the reviewer's order) form two pairs, each a criticism and the requests that follow from it. We answer W1 together with W3, then W2 together with W4, and number the quotes from each paragraph in the order we answer them (for example W1.2).
W1 and W3. The collapse and the exponent
The reviewer is right about what the collapse supports. The measurements in the submission stand: in the schedules we tested, a larger initial scale and a lower learning rate delay the decay, and the curves collapse approximately. What was not aligned was the wording, since "only act through this ratio" reads as a statement about every schedule, while the experiments covered one family. To resolve this we ran the checks the reviewer proposes and new experiments under other schedule shapes. They show where the relationship holds, and the revision states it there: an approximate collapse within the family of warmup schedules, with no exponent reported as a measured constant.
We narrow the claim to the family of warmup schedules. The submitted runs tested "only through the cumulative learning rate" within one family. The runs changed the multiplier of a single schedule, and inside its warmup the accumulated learning rate is proportional to multiplier × step², so they cannot tell the accumulated learning rate from other functionals of that schedule. We therefore trained networks under four further schedule shapes, at 14M and at 70M, from three initializations each: a constant learning rate from the first step, a 1,000-step warmup followed by a constant rate, a two-level schedule (a quarter of the peak until step 2,000, then the peak), and a high-then-low schedule (the peak until step 500, then a tenth of it).
Memory at equal accumulated learning rate (14M, mean of three initializations)
The accumulated learning rate is counted in steps at the peak rate.