INDEX

Explanations

words related to unauthorized or unacceptable actions or situations

oai_token-act-pair · gpt-3.5-turbo

terms related to validity, approval, and status of actions or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ioxide

-0.82

Clar

-0.72

zel

-0.71

otto

-0.70

ois

-0.69

alon

-0.69

frey

-0.66

udeb

-0.65

APH

-0.65

crit

-0.65

POSITIVE LOGITS

 distractions

0.77

adoes

0.74

 excuses

0.73

nesses

0.73

 anymore

0.70

 wastes

0.70

 pregnancies

0.69

 Abandon

0.65

ly

0.64

 Territory

0.63

Activations Density 0.123%

words related to unauthorized or unacceptable actions or situations

terms related to validity, approval, and status of actions or entities

No Comments

No Known Activations