INDEX

Explanations

references to the United States and its actions or characteristics

oai_token-act-pair · gpt-3.5-turbo

the word "is" used in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Telesc

-0.75

rouse

-0.71

idian

-0.66

ossom

-0.62

ees

-0.60

urry

-0.60

iates

-0.60

amount

-0.60

oso

-0.58

 Afterwards

-0.57

POSITIVE LOGITS

 indeed

0.95

 definitely

0.95

senal

0.91

 always

0.91

 certainly

0.89

 currently

0.87

now

0.87

 largely

0.86

 generally

0.85

 still

0.85

Activations Density 0.662%

references to the United States and its actions or characteristics

the word "is" used in various contexts

No Comments

No Known Activations