INDEX

Explanations

phrases related to information releases or statements

oai_token-act-pair · gpt-3.5-turbo

references to reports and statements, particularly those related to government or public releases

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Huss

-0.71

apesh

-0.69

gage

-0.65

Glass

-0.65

 Upgrade

-0.63

Joined

-0.63

connected

-0.63

witch

-0.63

FIELD

-0.61

ative

-0.61

POSITIVE LOGITS

 redacted

0.83

 transcripts

0.83

 information

0.78

 memoir

0.74

 transcript

0.73

 corrections

0.72

ebin

0.72

 apologies

0.70

 trove

0.70

 catalogue

0.70

Activations Density 0.209%

phrases related to information releases or statements

references to reports and statements, particularly those related to government or public releases

No Comments

No Known Activations