INDEX

Explanations

phrases related to repetition and continuous actions

oai_token-act-pair · gpt-3.5-turbo

patterns of repetition or redundancy in actions or phrases

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

izon

-0.57

adelphia

-0.53

rez

-0.53

anchester

-0.52

ibrary

-0.51

Ã¡s

-0.48

github

-0.45

bard

-0.45

 Marlins

-0.45

cture

-0.44

POSITIVE LOGITS

 alike

1.33

 respectively

0.99

 thereof

0.97

 accordingly

0.86

 thereafter

0.82

 thereto

0.80

 versa

0.69

 therein

0.69

depending

0.65

 afterwards

0.62

Activations Density 4.400%

phrases related to repetition and continuous actions

patterns of repetition or redundancy in actions or phrases

No Comments

No Known Activations