INDEX

Explanations

image-based descriptions mentioning elements or attributes

oai_token-act-pair · gpt-3.5-turbo

phrases that include the word "of"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

tesy

-0.86

initions

-0.84

States

-0.81

isons

-0.78

inations

-0.78

days

-0.74

uations

-0.73

ivals

-0.73

ilib

-0.72

bowl

-0.71

POSITIVE LOGITS

 flesh

0.97

 grass

0.96

 gray

0.96

 parchment

0.95

 coastline

0.92

 land

0.90

 cloth

0.90

 dough

0.87

 rubble

0.87

 wood

0.87

Activations Density 0.125%

image-based descriptions mentioning elements or attributes

phrases that include the word "of"

No Comments

No Known Activations