INDEX

Explanations

keywords related to performance evaluation and capabilities

oai_token-act-pair · gpt-3.5-turbo

references to performance metrics or evaluation results

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

cot

-0.74

alg

-0.70

shi

-0.69

vy

-0.69

bara

-0.68

ski

-0.66

Pg

-0.66

ship

-0.65

ke

-0.65

yg

-0.64

POSITIVE LOGITS

 performance

1.10

Performance

1.00

 Performance

0.96

 performances

0.91

 improvement

0.90

performance

0.90

 enhancement

0.89

 tuning

0.87

 Improvement

0.85

 optimizations

0.84

Activations Density 0.018%

keywords related to performance evaluation and capabilities

references to performance metrics or evaluation results

No Comments

No Known Activations