INDEX

Explanations

instances where the text mentions contrasting or different options

oai_token-act-pair · gpt-3.5-turbo

references to alternative options or consequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Encyclopedia

-0.70

 Mehran

-0.64

 Abstract

-0.63

ãĥī

-0.61

UES

-0.60

Lenin

-0.60

oret

-0.58

 Reef

-0.58

Upload

-0.58

forestation

-0.57

POSITIVE LOGITS

worldly

1.19

 besides

0.94

 entirely

0.78

arettes

0.73

where

0.70

isin

0.70

Joined

0.69

mia

0.68

adin

0.68

 Â¯

0.64

Activations Density 0.037%

instances where the text mentions contrasting or different options

references to alternative options or consequences

No Comments

No Known Activations