INDEX

Explanations

phrases related to positions or assertions

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate claims or statements made in a formal context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-0.70

FTWARE

-0.66

 Mages

-0.63

 bandwagon

-0.61

illion

-0.60

}}

-0.59

)].

-0.58

 Folder

-0.57

 bleach

-0.57

 rename

-0.57

POSITIVE LOGITS

processing

0.75

conduct

0.66

lists

0.65

oug

0.65

dk

0.65

©¶æ

0.64

conv

0.64

jug

0.63

Ĭ±

0.62

anced

0.62

Activations Density 0.248%

phrases related to positions or assertions

phrases that indicate claims or statements made in a formal context

No Comments

No Known Activations