INDEX

Explanations

terms related to cooking

oai_token-act-pair · gpt-3.5-turbo

references to cooking-related terms and processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

idas

-0.77

emale

-0.71

eson

-0.68

umbn

-0.68

orporated

-0.68

orp

-0.66

orb

-0.65

hovah

-0.64

 Baghd

-0.64

upon

-0.63

POSITIVE LOGITS

 cooks

0.97

 cook

0.93

stall

0.93

 stove

0.93

fry

0.92

cook

0.91

Recipe

0.89

 oven

0.88

 meals

0.88

Cook

0.87

Activations Density 0.017%

terms related to cooking

references to cooking-related terms and processes

No Comments

No Known Activations