INDEX

Explanations

proper nouns, likely names of organizations or entities

oai_token-act-pair · gpt-3.5-turbo

references to news organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ront

-0.78

 mattered

-0.65

ificial

-0.57

akens

-0.56

ktop

-0.55

ezvous

-0.54

odied

-0.53

zees

-0.53

 confront

-0.53

oward

-0.51

POSITIVE LOGITS

 datas

0.85

0.83

:[

0.83

0.82

,[

0.82

,,

0.79

*,

0.73

 docs

0.72

%,

0.72

 estimates

0.71

Activations Density 0.202%

proper nouns, likely names of organizations or entities

references to news organizations

No Comments

No Known Activations