INDEX

Explanations

terms related to news stories and articles

oai_token-act-pair · gpt-3.5-turbo

instances of the word "Story" that signal a continuation in a document

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

ŃĶ

-0.74

»Ĵ

-0.69

«ĺ

-0.68

ergic

-0.68

agles

-0.67

agall

-0.65

anguage

-0.63

 fame

-0.63

LIA

-0.63

ullivan

-0.62

POSITIVE LOGITS

telling

0.97

 Continued

0.89

 highlights

0.76

line

0.74

 Transcript

0.72

tenance

0.70

stal

0.70

 Contin

0.69

 Highlights

0.68

 Details

0.68

Activations Density 0.022%

terms related to news stories and articles

instances of the word "Story" that signal a continuation in a document

No Comments

No Known Activations