INDEX

Explanations

locations or places

oai_token-act-pair · gpt-3.5-turbo

proper nouns and specific names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 conflic

-0.77

ccording

-0.71

 unden

-0.70

ãĤ¨

-0.64

 citiz

-0.59

etheless

-0.58

ongyang

-0.58

 yours

-0.58

ĺ

-0.57

 exha

-0.57

POSITIVE LOGITS

sburg

0.79

 Chronicle

0.61

shire

0.60

ridge

0.58

 Crusade

0.57

woods

0.57

ite

0.56

 Chronicles

0.56

wine

0.56

asca

0.54

Activations Density 0.781%

locations or places

proper nouns and specific names

No Comments

No Known Activations