INDEX

Explanations

phrases indicating something is not socially or morally appropriate

oai_token-act-pair · gpt-3.5-turbo

words related to acceptability or unacceptability

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ynthesis

-0.83

ocket

-0.80

frey

-0.76

planes

-0.75

ilant

-0.73

lets

-0.72

dream

-0.72

 helic

-0.69

berry

-0.69

wright

-0.68

POSITIVE LOGITS

GoldMagikarp

0.80

 srfAttach

0.75

CPC

0.75

lihood

0.73

 deviations

0.72

ible

0.71

âĶĢâĶĢ

0.70

 precedent

0.69

itable

0.69

Gi

0.68

Activations Density 0.029%

phrases indicating something is not socially or morally appropriate

words related to acceptability or unacceptability

No Comments

No Known Activations