INDEX

Explanations

phrases indicating distinguishing characteristics or unique features

oai_token-act-pair · gpt-3.5-turbo

phrases that identify distinguishing characteristics or qualities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

endix

-0.80

lance

-0.75

lex

-0.74

ixon

-0.72

xon

-0.70

erenn

-0.69

odder

-0.67

rentice

-0.67

tch

-0.66

cow

-0.66

POSITIVE LOGITS

 them

0.71

orno

0.69

Cu

0.69

rament

0.67

 distinguishes

0.67

 humanity

0.67

 Sanct

0.66

ably

0.64

 these

0.63

 Tyrann

0.63

Activations Density 0.107%

phrases indicating distinguishing characteristics or unique features

phrases that identify distinguishing characteristics or qualities

No Comments

No Known Activations