INDEX

Explanations

phrases indicating a large quantity or proportion of something

oai_token-act-pair · gpt-3.5-turbo

terms indicating large quantities or representations in a group

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

ollah

-0.68

 nearest

-0.67

imir

-0.63

pson

-0.62

 elig

-0.61

 slump

-0.59

 Centauri

-0.59

 superintendent

-0.59

 pastor

-0.58

istered

-0.58

POSITIVE LOGITS

(>

1.11

(~

0.79

($)

0.71

UID

0.67

atana

0.66

hyde

0.66

ersion

0.66

ANCE

0.65

ixel

0.65

IU

0.63

Activations Density 0.088%

phrases indicating a large quantity or proportion of something

terms indicating large quantities or representations in a group

No Comments

No Known Activations