INDEX

Explanations

statements or declarations of importance

oai_token-act-pair · gpt-3.5-turbo

mentions of the term 'statement' in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.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

rys

-0.79

NetMessage

-0.77

ikarp

-0.76

bid

-0.76

rowd

-0.76

elsius

-0.75

sites

-0.73

cffff

-0.71

Friend

-0.70

bat

-0.69

POSITIVE LOGITS

 statements

0.94

 regarding

0.89

 statement

0.88

 unequivocally

0.84

 uttered

0.83

 condemning

0.81

ariat

0.80

 attributed

0.79

 denouncing

0.79

 announcing

0.79

Activations Density 0.054%

statements or declarations of importance

mentions of the term 'statement' in various contexts

No Comments

No Known Activations