INDEX

Explanations

phrases indicating changing locations or positions

oai_token-act-pair · gpt-3.5-turbo

phrases indicating movement or transition

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

glass

-0.69

BALL

-0.66

Glass

-0.65

TD

-0.63

 indu

-0.63

psi

-0.62

urable

-0.62

 snapped

-0.61

cone

-0.60

 Statistical

-0.59

POSITIVE LOGITS

 obscurity

0.74

ilet

0.74

 await

0.71

ontent

0.71

Sad

0.70

bia

0.69

 pursue

0.66

arettes

0.66

 fend

0.66

 Dign

0.65

Activations Density 0.195%

phrases indicating changing locations or positions

phrases indicating movement or transition

No Comments

No Known Activations