INDEX

Explanations

technical terms related to software development and progress updates

oai_token-act-pair · gpt-3.5-turbo

statements about personal experiences or announcements

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

 disproportionately

-0.68

 Saddam

-0.68

 ordinarily

-0.68

 stereotyp

-0.67

 locality

-0.66

 sheltered

-0.66

 centr

-0.65

usterity

-0.65

 whites

-0.64

 traditionally

-0.64

POSITIVE LOGITS

âĢ

1.00

 ðŁĳ

0.99

[/

0.96

âľ

0.96

 ðŁĻĤ

0.96

 âĿ

0.93

 THANK

0.91

 âľ

0.90

Insp

0.89

 Kickstarter

0.89

Activations Density 0.700%

technical terms related to software development and progress updates

statements about personal experiences or announcements

No Comments

No Known Activations