INDEX

Explanations

statements or notes related to information or updates

oai_token-act-pair · gpt-3.5-turbo

assertive statements and personal opinions

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

tnc

-0.78

rouse

-0.71

anooga

-0.70

 manageable

-0.66

ieu

-0.66

rift

-0.64

 overwhelm

-0.64

icking

-0.64

ickers

-0.63

akable

-0.63

POSITIVE LOGITS

 NEVER

1.35

 ALSO

1.29

 actually

1.18

 ONLY

1.18

 ALWAYS

1.11

DID

1.05

NOT

1.04

 REALLY

0.98

 DOES

0.98

 never

0.90

Activations Density 0.519%

statements or notes related to information or updates

assertive statements and personal opinions

No Comments

No Known Activations