INDEX

Explanations

phrases related to various reasons and explanations

oai_token-act-pair · gpt-3.5-turbo

references to significant reasons or explanations 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 @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

IELD

-0.66

taker

-0.62

owers

-0.54

riage

-0.54

ERROR

-0.54

ocl

-0.53

exit

-0.53

OWER

-0.53

å§

-0.53

 keeper

-0.53

POSITIVE LOGITS

 ranging

1.61

 include

1.41

 varied

1.40

 ranged

1.33

 includ

1.31

 including

1.24

including

1.23

 summarized

1.21

 varying

1.19

ranging

1.15

Activations Density 0.882%

phrases related to various reasons and explanations

references to significant reasons or explanations within the text

No Comments

No Known Activations