INDEX

Explanations

phrases related to notable or important things

oai_token-act-pair · gpt-3.5-turbo

the word "notable" and variations of it

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

©¶æ

-0.83

claimer

-0.81

prep

-0.75

alyst

-0.74

anguage

-0.73

nan

-0.72

culus

-0.70

nery

-0.70

cock

-0.68

vacc

-0.68

POSITIVE LOGITS

Notable

0.91

 landmarks

0.91

 exceptions

0.90

 exception

0.89

 milestones

0.84

 accomplishments

0.81

 notable

0.80

itarian

0.79

 feats

0.78

 feat

0.76

Activations Density 0.036%

phrases related to notable or important things

the word "notable" and variations of it

No Comments

No Known Activations