INDEX

Explanations

pronouns referring to unspecified entities

oai_token-act-pair · gpt-3.5-turbo

references to entities or objects being discussed in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

âĢ¢âĢ¢

-0.74

mire

-0.69

âĺħâĺħ

-0.68

Iowa

-0.68

odge

-0.67

amer

-0.67

Party

-0.66

SPA

-0.66

ILE

-0.64

 ORDER

-0.63

POSITIVE LOGITS

atically

1.29

 selves

1.29

selves

1.26

atic

1.14

self

0.96

 conduc

0.83

atics

0.82

 behav

0.73

MpServer

0.72

 orally

0.72

Activations Density 0.155%

pronouns referring to unspecified entities

references to entities or objects being discussed in the text

No Comments

No Known Activations