INDEX

Explanations

words related to escaping or fleeing

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "escape" in various contexts

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

sonian

-0.84

inki

-0.73

ammy

-0.71

trust

-0.71

eous

-0.69

iop

-0.67

older

-0.67

urally

-0.66

alky

-0.66

aic

-0.66

POSITIVE LOGITS

 escape

1.03

 escapes

0.93

 escaped

0.89

 Torment

0.82

escape

0.82

 escaping

0.78

 hatch

0.76

 detection

0.74

APE

0.74

 convict

0.71

Activations Density 0.009%

words related to escaping or fleeing

mentions of the word "escape" in various contexts

No Comments

No Known Activations