INDEX

Explanations

words related to technical processes or actions

oai_token-act-pair · gpt-3.5-turbo

end-of-document indicators and non-content tokens

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 destro

-0.70

 respectively

-0.59

$$$$

-0.53

jri

-0.53

 âĢ¦"

-0.53

 challeng

-0.53

 disg

-0.51

 thereof

-0.51

 thereto

-0.50

Ò

-0.48

POSITIVE LOGITS

 Profile

0.78

rimination

0.56

 Warranty

0.55

 Edit

0.52

 Background

0.52

 Description

0.51

 Picks

0.51

 Removal

0.51

 Releases

0.51

pedia

0.51

Activations Density 0.563%

words related to technical processes or actions

end-of-document indicators and non-content tokens

No Comments

No Known Activations