INDEX

Explanations

words related to laws, regulations, policies, and governmental actions

oai_token-act-pair · gpt-3.5-turbo

phrases indicating existence or presence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

maxwell

-0.68

lex

-0.65

avier

-0.64

ares

-0.62

ival

-0.62

 rethink

-0.62

lyak

-0.59

*/

-0.59

âĢ¢

-0.58

Stats

-0.57

POSITIVE LOGITS

 plag

1.27

 hallmark

0.93

 emanating

0.91

 surround

0.87

 engulf

0.87

 underpin

0.84

 inherent

0.83

 unfolding

0.82

 endemic

0.82

 prevalent

0.81

Activations Density 0.181%

words related to laws, regulations, policies, and governmental actions

phrases indicating existence or presence

No Comments

No Known Activations