INDEX

Explanations

phrases related to rules, regulations, or restrictions

oai_token-act-pair · gpt-3.5-turbo

terms related to restrictions or prohibitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

eon

-0.83

eah

-0.74

itched

-0.69

 Balkans

-0.68

neau

-0.66

lycer

-0.66

tops

-0.65

ilant

-0.62

ipel

-0.62

tone

-0.61

POSITIVE LOGITS

 tampering

0.78

untarily

0.77

idding

0.75

 gambling

0.74

 smoking

0.74

bidden

0.74

 substances

0.72

 misuse

0.69

 prohibited

0.69

¿½

0.68

Activations Density 0.056%

phrases related to rules, regulations, or restrictions

terms related to restrictions or prohibitions

No Comments

No Known Activations