INDEX

Explanations

phrases related to totality or completeness

oai_token-act-pair · gpt-3.5-turbo

instances of the word "entire" used with varying contexts

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

friends

-0.73

bid

-0.71

ju

-0.69

rants

-0.66

killers

-0.65

 mates

-0.65

kick

-0.64

akers

-0.63

cru

-0.63

zy

-0.62

POSITIVE LOGITS

 spectrum

0.79

 ordeal

0.75

 length

0.75

 arsenal

0.75

 globe

0.75

gam

0.75

heartedly

0.74

hearted

0.74

 ecosystem

0.73

 playthrough

0.72

Activations Density 0.023%

phrases related to totality or completeness

instances of the word "entire" used with varying contexts

No Comments

No Known Activations