INDEX

Explanations

mentions of death or related terms

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "death."

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Available

-0.79

Cola

-0.74

ĸļ

-0.74

Ĭ±

-0.74

Alpha

-0.71

Avg

-0.70

 Available

-0.68

OPER

-0.67

ĳ

-0.67

 Collider

-0.67

POSITIVE LOGITS

blow

0.89

stroke

0.87

death

0.86

touch

0.86

anguage

0.86

guards

0.84

 death

0.83

hound

0.83

match

0.82

fighter

0.81

Activations Density 0.021%

mentions of death or related terms

mentions of the word "death."

No Comments

No Known Activations