INDEX

Explanations

phrases related to dangerous or risky situations

oai_token-act-pair · gpt-3.5-turbo

phrases related to dangerous or risky scenarios

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rotein

-0.88

sonian

-0.83

leaf

-0.77

tool

-0.73

 Bones

-0.72

cott

-0.71

track

-0.71

rik

-0.69

uv

-0.68

artisan

-0.67

POSITIVE LOGITS

 situations

1.06

uations

0.95

 involving

0.94

 arise

0.80

 Situation

0.79

 circumstances

0.79

uation

0.78

 scenarios

0.73

 arisen

0.73

ually

0.72

Activations Density 0.021%

phrases related to dangerous or risky situations

phrases related to dangerous or risky scenarios

No Comments

No Known Activations