INDEX

Explanations

prepositions followed by locations or figures

oai_token-act-pair · gpt-3.5-turbo

Triggered by @bot

the word "in" and its context, indicating its focus on spatial or situational elements within the text

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_post

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

member

-0.58

iper

-0.57

onom

-0.56

oran

-0.54

 escorted

-0.53

 impaired

-0.52

 pals

-0.51

ills

-0.51

 assisted

-0.50

nell

-0.49

POSITIVE LOGITS

between

1.43

 between

1.04

effic

1.02

clusions

0.96

efficiency

0.92

animate

0.91

 existence

0.88

 sight

0.85

 there

0.83

 relation

0.82

Activations Density 0.237%

prepositions followed by locations or figures

the word "in" and its context, indicating its focus on spatial or situational elements within the text

No Known Activations