INDEX

Explanations

terms related to physical scarring

oai_token-act-pair · gpt-3.5-turbo

references to scar tissue or scars

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

£ı

-0.84

gaard

-0.80

hower

-0.77

ablishment

-0.73

ullivan

-0.72

ostics

-0.72

eger

-0.69

ĨĴ

-0.66

aminer

-0.66

Ģ

-0.65

POSITIVE LOGITS

red

1.04

ring

0.94

fing

0.94

lets

0.91

crow

0.90

pered

0.88

ves

0.87

lett

0.84

pees

0.83

face

0.83

Activations Density 0.012%

terms related to physical scarring

references to scar tissue or scars

No Comments

No Known Activations