INDEX

Explanations

proper nouns or specific entities

oai_token-act-pair · gpt-3.5-turbo

instances of the verb "have" in various grammatical contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

arter

-0.71

 apology

-0.62

ocol

-0.60

 Apart

-0.60

 Recap

-0.59

 Anyway

-0.59

 ÂłÂł

-0.58

********

-0.57

 dispatch

-0.57

 Saying

-0.57

POSITIVE LOGITS

been

1.31

 been

1.30

 undergone

1.06

 Been

0.98

 endured

0.96

 arisen

0.95

 become

0.94

kell

0.90

 existed

0.89

 worked

0.88

Activations Density 0.181%

proper nouns or specific entities

instances of the verb "have" in various grammatical contexts

No Comments

No Known Activations