INDEX

Explanations

TV show or movie titles

oai_token-act-pair · gpt-3.5-turbo

closed parentheses or the end of statements within lists

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

 answ

-0.97

omorphic

-0.74

itory

-0.71

 compan

-0.71

spo

-0.69

natureconservancy

-0.69

ocial

-0.69

 tremend

-0.67

uga

-0.66

 commer

-0.65

POSITIVE LOGITS

 Committees

0.75

 Frames

0.74

ãĥİ

0.74

)].

0.72

>]

0.71

 Syndicate

0.69

 âĵĺ

0.67

])

0.67

arthy

0.66

 Parables

0.66

Activations Density 0.200%

TV show or movie titles

closed parentheses or the end of statements within lists

No Comments

No Known Activations