INDEX

Explanations

proper nouns or names used for comparisons

oai_token-act-pair · gpt-3.5-turbo

phrases involving the word "likes" and its context

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

INTON

-0.63

ISH

-0.62

å£«

-0.62

 Procedures

-0.61

 Donation

-0.59

 Sources

-0.59

 sources

-0.56

cas

-0.56

NING

-0.55

 Springs

-0.55

POSITIVE LOGITS

liest

1.14

lihood

1.06

hots

0.81

liness

0.78

creen

0.77

lier

0.77

wikipedia

0.74

paces

0.72

of

0.70

minded

0.69

Activations Density 0.027%

proper nouns or names used for comparisons

phrases involving the word "likes" and its context

No Comments

No Known Activations