INDEX

Explanations

phrases related to changes or edits in various contexts

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "revision" and its variations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Flu

-0.71

eu

-0.66

ħ

-0.65

zer

-0.65

Rh

-0.65

ells

-0.64

resp

-0.64

 Skies

-0.63

orld

-0.62

Ability

-0.62

POSITIVE LOGITS

 revision

1.45

 revisions

1.07

 guiActiveUn

1.03

 Revision

1.00

uberty

0.82

Lenin

0.80

itized

0.80

udicrous

0.79

itial

0.78

ismo

0.75

Activations Density 0.008%

phrases related to changes or edits in various contexts

mentions of the word "revision" and its variations

No Comments

No Known Activations