INDEX

Explanations

phrases mentioning signs or indications

oai_token-act-pair · gpt-3.5-turbo

terms related to signs or indicators of various situations or conditions

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

ancies

-0.72

ichen

-0.64

elta

-0.64

 livest

-0.62

asus

-0.61

quer

-0.60

onut

-0.59

ecycle

-0.59

 accomp

-0.58

tones

-0.57

POSITIVE LOGITS

of

0.96

 that

0.86

 thereof

0.84

 indicating

0.78

 indicative

0.76

ifier

0.75

posts

0.68

post

0.68

 indication

0.66

atory

0.63

Activations Density 0.088%

phrases mentioning signs or indications

terms related to signs or indicators of various situations or conditions

No Comments

No Known Activations