INDEX

Explanations

phrases indicating someone expressing an opinion or belief

oai_token-act-pair · gpt-3.5-turbo

statements indicating personal or institutional positions and declarations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ctors

-0.81

OTAL

-0.63

ername

-0.62

 Sabha

-0.62

quin

-0.62

JV

-0.61

 Combine

-0.59

Commerce

-0.59

hoff

-0.58

EStreamFrame

-0.57

POSITIVE LOGITS

 repeatedly

1.20

 publicly

0.91

 since

0.86

 unequivocally

0.85

avering

0.83

 reluctance

0.82

 contingency

0.81

 numerous

0.80

 intentions

0.79

 lately

0.78

Activations Density 0.193%

phrases indicating someone expressing an opinion or belief

statements indicating personal or institutional positions and declarations

No Comments

No Known Activations