INDEX

Explanations

comma characters

oai_token-act-pair · gpt-3.5-turbo

occurrences of commas 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

 advis

-0.65

 exha

-0.63

amorph

-0.62

 vulner

-0.62

Tokens

-0.62

 behavi

-0.61

doors

-0.61

 corrid

-0.60

wagen

-0.58

accompan

-0.57

POSITIVE LOGITS

actionDate

0.81

mosp

0.80

Psy

0.72

lement

0.63

 ILCS

0.61

hov

0.61

 Entered

0.60

taboola

0.59

nee

0.59

 Ages

0.59

Activations Density 0.079%

comma characters

occurrences of commas in the text

No Comments

No Known Activations