INDEX

Explanations

phrases related to specific geographical locations, particularly those with the word "Field"

oai_token-act-pair · gpt-3.5-turbo

references to specific sports fields or stadiums

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

mx

-0.63

 âĶľ

-0.63

 trusts

-0.62

 triglycer

-0.62

lu

-0.61

 booze

-0.61

tails

-0.60

ĻĤ

-0.59

 euro

-0.59

 prox

-0.59

POSITIVE LOGITS

 Field

4.03

Field

2.65

 field

2.38

 Fields

2.13

field

1.96

 fields

1.78

FIELD

1.70

fields

1.59

 Stadium

1.16

 fielding

1.08

Activations Density 0.016%

phrases related to specific geographical locations, particularly those with the word "Field"

references to specific sports fields or stadiums

No Comments

No Known Activations