INDEX

Explanations

phrases related to equality and inclusivity

oai_token-act-pair · gpt-3.5-turbo

statements emphasizing equality and inclusivity among people

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

vernment

-0.69

qus

-0.66

Newsletter

-0.65

edia

-0.62

ahime

-0.61

\<

-0.59

few

-0.58

 Rather

-0.57

Ö¼

-0.57

interstitial

-0.56

POSITIVE LOGITS

 except

1.11

 equally

0.89

 alike

0.87

 interconnected

0.84

ratulations

0.83

 equal

0.81

equal

0.79

 converge

0.77

except

0.74

 together

0.74

Activations Density 0.437%

phrases related to equality and inclusivity

statements emphasizing equality and inclusivity among people

No Comments

No Known Activations