INDEX

Explanations

phrases indicating a significant event or development

oai_token-act-pair · gpt-3.5-turbo

instances of the word "what" related to descriptions, labels, or definitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 crying

-0.58

 largeDownload

-0.57

 Emirates

-0.56

 inserting

-0.55

enance

-0.55

 stride

-0.54

strap

-0.54

enger

-0.53

robe

-0.53

 aster

-0.53

POSITIVE LOGITS

soever

1.07

 transpired

0.96

 happens

0.96

 happened

0.93

 constitutes

0.90

 amounted

0.83

orate

0.83

 constituted

0.79

atus

0.78

 awaits

0.73

Activations Density 0.083%

phrases indicating a significant event or development

instances of the word "what" related to descriptions, labels, or definitions

No Comments

No Known Activations