INDEX

Explanations

pronouns and related words denoting people or groups

oai_token-act-pair · gpt-3.5-turbo

references to specific individuals or pronouns indicating their actions

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

é¾įåĸļå£«

-0.83

çĶ°

-0.76

inence

-0.74

Sense

-0.74

-0.73

 Defin

-0.72

MENT

-0.70

 mosqu

-0.70

cies

-0.69

ãĥĥãĥī

-0.68

POSITIVE LOGITS

 Beckham

0.62

 Ajax

0.62

rites

0.59

 consent

0.59

/"

0.58

 Kinect

0.58

 Canaver

0.57

 Couch

0.57

 Franco

0.57

platform

0.57

Activations Density 0.000%

pronouns and related words denoting people or groups

references to specific individuals or pronouns indicating their actions

No Comments

No Known Activations