INDEX

Explanations

phrases or sentences emphasizing discussions or concerns

oai_token-act-pair · gpt-3.5-turbo

topics of discussion or focus within a conversation

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

 disabled

-0.54

 satisfaction

-0.53

 priesthood

-0.53

cknowled

-0.53

Tol

-0.52

 Admir

-0.52

 Priest

-0.51

 Quentin

-0.51

 pacif

-0.51

 needy

-0.50

POSITIVE LOGITS

 during

0.80

ãĢĤ

0.75

.:

0.74

abouts

0.72

.,

0.71

 beforehand

0.71

.(

0.70

–

0.69

 anyways

0.68

Activations Density 0.249%

phrases or sentences emphasizing discussions or concerns

topics of discussion or focus within a conversation

No Comments

No Known Activations