INDEX

Explanations

numerical information and statistical estimates

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

Data Type

torch.float32

Hook Point

blocks.8.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

xtap

-0.90

ãĥĩ

-0.83

 Himself

-0.80

gur

-0.78

ãĤ´ãĥ³

-0.70

Ã±

-0.68

ãĥ¬

-0.66

EP

-0.66

ãĥĵ

-0.65

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.64

POSITIVE LOGITS

 otherwise

0.75

 probable

0.72

 they

0.68

 discrepancies

0.68

 irregularities

0.68

 there

0.66

 that

0.66

 delays

0.65

 rampant

0.65

 anecd

0.63

Activations Density 2.652%

numerical information and statistical estimates

No Comments

No Known Activations