INDEX

Explanations

words related to injuries and wounds

oai_token-act-pair · gpt-3.5-turbo

references to injuries or wounds

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

anooga

-0.79

alty

-0.76

acons

-0.74

*/(

-0.71

poons

-0.70

hist

-0.68

ordan

-0.67

agically

-0.67

authorized

-0.67

ansk

-0.65

POSITIVE LOGITS

 wound

0.96

 wounds

0.93

robe

0.83

burn

0.78

0.74

 Whedon

0.74

bags

0.72

bolt

0.70

itial

0.70

 inflicted

0.69

Activations Density 0.011%

words related to injuries and wounds

references to injuries or wounds

No Comments

No Known Activations