INDEX

Explanations

parentheses used to provide additional or explanatory information

oai_token-act-pair · gpt-3.5-turbo

parentheses and their usage in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

orn

-0.72

lav

-0.72

quished

-0.70

 Leilan

-0.68

($)

-0.68

 Taiwan

-0.64

 benefic

-0.64

 veget

-0.63

Rye

-0.62

ob

-0.61

POSITIVE LOGITS

sic

1.42

laughs

1.08

emphasis

0.99

âĢ¦)

0.95

during

0.91

Wednesday

0.88

when

0.87

Tuesday

0.87

said

0.86

Monday

0.84

Activations Density 0.078%

parentheses used to provide additional or explanatory information

parentheses and their usage in the text

No Comments

No Known Activations