INDEX

Explanations

phrases related to legal, political, or administrative contexts

oai_token-act-pair · gpt-3.5-turbo

punctuation marks, specifically commas

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

iple

-0.64

¬¼

-0.61

natureconservancy

-0.59

 corrid

-0.57

ety

-0.53

heast

-0.49

 murd

-0.47

©

-0.47

izon

-0.47

 metic

-0.46

POSITIVE LOGITS

 albeit

0.77

etc

0.76

 although

0.74

 however

0.74

 namely

0.73

 respectively

0.68

huh

0.64

 though

0.64

 whereas

0.62

 meanwhile

0.61

Activations Density 1.448%

phrases related to legal, political, or administrative contexts

punctuation marks, specifically commas

No Comments

No Known Activations