INDEX

Explanations

quotations

oai_token-act-pair · gpt-3.5-turbo

quoted speech or dialogue

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

 Salem

-0.68

TOM

-0.65

 Bronx

-0.65

 Chennai

-0.65

SH

-0.64

 Potato

-0.64

 Akira

-0.63

 Gors

-0.63

 Luck

-0.63

 Dispatch

-0.63

POSITIVE LOGITS

âĢ¦]

1.40

...]

1.33

english

1.06

REDACTED

0.88

ftime

0.86

ety

0.82

][

0.78

']

0.76

!]

0.73

ideshow

0.73

Activations Density 0.028%

quotations

quoted speech or dialogue

No Comments

No Known Activations