INDEX

Explanations

references to people's names

oai_token-act-pair · gpt-3.5-turbo

occurrences of possessive pronouns related to people or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

Zi

-0.68

 Izan

-0.66

Kou

-0.65

ylon

-0.65

 Iranians

-0.63

 Europeans

-0.62

Cel

-0.61

 Babe

-0.61

otin

-0.60

 Lerner

-0.60

POSITIVE LOGITS

own

1.05

 tremend

0.86

 briefs

0.76

Own

0.75

yssey

0.75

 introductory

0.74

 travels

0.73

Ĥİ

0.70

 stride

0.69

panic

0.68

Activations Density 0.132%

references to people's names

occurrences of possessive pronouns related to people or entities

No Comments

No Known Activations