INDEX

Explanations

references to the English soccer coach and former player Harry Redknapp

oai_token-act-pair · gpt-3.5-turbo

variations of the word "red" and its associations in different contexts

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

hran

-0.76

schild

-0.72

sonian

-0.71

IAS

-0.61

BIL

-0.60

 Sabha

-0.58

Ö¼

-0.58

LM

-0.57

KY

-0.54

gol

-0.54

POSITIVE LOGITS

oxide

0.78

Ń·

0.63

vertisement

0.63

reme

0.61

asion

0.59

 Howell

0.58

Grey

0.57

ĪĴ

0.57

 Tide

0.56

irted

0.55

Activations Density 0.094%

references to the English soccer coach and former player Harry Redknapp

variations of the word "red" and its associations in different contexts

No Comments

No Known Activations