INDEX

Explanations

mentions of things being the "first" in a specific context

oai_token-act-pair · gpt-3.5-turbo

references to pioneering studies or groundbreaking cases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Favorite

-0.71

mar

-0.69

gar

-0.69

nick

-0.65

hun

-0.62

beaut

-0.61

 sucks

-0.60

blade

-0.60

Weak

-0.59

asty

-0.59

POSITIVE LOGITS

 EVER

0.94

 ever

0.91

ever

0.90

 openly

0.86

 publicly

0.83

 formally

0.83

 foray

0.81

 officially

0.80

OPLE

0.78

 remotely

0.77

Activations Density 0.240%

mentions of things being the "first" in a specific context

references to pioneering studies or groundbreaking cases

No Comments

No Known Activations