INDEX

Explanations

French names or words

oai_token-act-pair · gpt-3.5-turbo

specific character sequences or key strings 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 @ 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

yip

-0.68

jriwal

-0.63

 merit

-0.57

boo

-0.56

 Wanted

-0.56

mates

-0.55

oteric

-0.55

 mood

-0.54

minster

-0.54

mate

-0.53

POSITIVE LOGITS

Ã®

0.91

issance

0.82

quin

0.79

uner

0.79

Ã¨re

0.77

eger

0.77

ean

0.74

ilton

0.74

ires

0.73

irs

0.73

Activations Density 0.148%

French names or words

specific character sequences or key strings in the text

No Comments

No Known Activations