INDEX

Explanations

words related to blaming or assigning responsibility

oai_token-act-pair · gpt-3.5-turbo

references to blame and shame in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

keyes

-0.82

 stocking

-0.66

 receptive

-0.64

 antidepressant

-0.63

İĭ

-0.63

aido

-0.63

 combust

-0.62

shr

-0.61

 intraven

-0.61

 tremend

-0.60

POSITIVE LOGITS

ame

1.21

ames

0.94

 Shame

0.82

 Chomsky

0.77

eus

0.77

ilial

0.75

ffen

0.72

leon

0.72

enment

0.69

Ã©s

0.68

Activations Density 0.008%

words related to blaming or assigning responsibility

references to blame and shame in various contexts

No Comments

No Known Activations