INDEX

Explanations

specific locations or names of places, potentially related to geographical areas

oai_token-act-pair · gpt-3.5-turbo

references to specific names and organizations

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

otype

-0.82

 Kubrick

-0.73

venants

-0.69

otypes

-0.68

quez

-0.66

icles

-0.64

 proc

-0.63

ctic

-0.63

helle

-0.63

POSITIVE LOGITS

lash

0.82

wash

0.78

fing

0.77

roots

0.76

Pwr

0.76

ande

0.75

lance

0.75

flies

0.73

trak

0.72

endish

0.72

Activations Density 0.069%

specific locations or names of places, potentially related to geographical areas

references to specific names and organizations

No Comments

No Known Activations