INDEX

Explanations

negative responses or refusals

oai_token-act-pair · gpt-3.5-turbo

negative affirmations or words expressing refusal

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

RAFT

-0.76

lycer

-0.75

iership

-0.74

ulative

-0.66

romeda

-0.66

endish

-0.65

IUM

-0.64

assies

-0.64

ual

-0.64

rious

-0.63

POSITIVE LOGITS

xious

1.12

zzle

0.99

 matter

0.93

except

0.92

obs

0.88

 longer

0.86

oses

0.86

ct

0.84

ises

0.83

AH

0.81

Activations Density 0.087%

negative responses or refusals

negative affirmations or words expressing refusal

No Comments

No Known Activations