INDEX

Explanations

statistical references and percentages related to various topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

venge

-0.74

jri

-0.70

Redd

-0.70

kefeller

-0.66

vised

-0.65

ãĤ´ãĥ³

-0.64

eworld

-0.61

 riches

-0.60

laden

-0.60

ding

-0.59

POSITIVE LOGITS

 respondents

0.89

 percent

0.86

percent

0.72

 Australians

0.71

 Americans

0.70

 Britons

0.70

of

0.70

 fewer

0.69

0.68

 Percent

0.68

Activations Density 0.042%

statistical references and percentages related to various topics

No Comments

No Known Activations