INDEX

Explanations

pronouns and verbs referring to actions or events

oai_token-act-pair · gpt-3.5-turbo

instances of the word "It" that indicate the subject of various sentences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

reference

-0.72

aldehyde

-0.66

 paran

-0.60

 Bunker

-0.59

oller

-0.59

%%%%

-0.58

IENT

-0.57

 welf

-0.56

gress

-0.56

 Offline

-0.55

POSITIVE LOGITS

 wasn

1.12

 hurts

1.07

 mattered

1.03

 hasn

1.03

zik

1.02

 seems

1.00

beh

0.99

ain

0.99

isn

0.98

 feels

0.97

Activations Density 0.163%

pronouns and verbs referring to actions or events

instances of the word "It" that indicate the subject of various sentences

No Comments

No Known Activations