INDEX

Explanations

mentions of the website "Snopes" and related content

oai_token-act-pair · gpt-3.5-turbo

mentions of "opes" and related terms, likely focusing on media or commentary about "snopes" and urban legends

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

draft

-0.84

sent

-0.78

mine

-0.76

chest

-0.72

à¨

-0.71

ston

-0.70

pron

-0.70

knit

-0.69

birds

-0.68

bird

-0.68

POSITIVE LOGITS

opes

1.54

ope

0.87

 eleph

0.84

cale

0.82

terday

0.80

agascar

0.79

heric

0.78

utterstock

0.76

 Aires

0.74

oglu

0.74

Activations Density 0.008%

mentions of the website "Snopes" and related content

mentions of "opes" and related terms, likely focusing on media or commentary about "snopes" and urban legends

No Comments

No Known Activations