INDEX

Explanations

negative language or criticism

oai_token-act-pair · gpt-3.5-turbo

negative expressions or critiques regarding various subjects

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

isations

-0.72

ATIONAL

-0.69

nda

-0.67

uclear

-0.67

ATIONS

-0.66

isation

-0.65

 rall

-0.62

tis

-0.62

Decl

-0.61

iferation

-0.61

POSITIVE LOGITS

asses

1.01

 dick

0.98

eries

0.97

sylvania

0.89

driver

0.89

shit

0.87

loads

0.82

bowl

0.80

tails

0.80

unin

0.78

Activations Density 0.016%

negative language or criticism

negative expressions or critiques regarding various subjects

No Comments

No Known Activations