INDEX

Explanations

mentions of similarities or commonalities between two entities

oai_token-act-pair · gpt-3.5-turbo

instances of the word "Both."

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

ļéĨĴ

-0.86

renheit

-0.80

fecture

-0.74

ANY

-0.72

istle

-0.71

Enlarge

-0.68

cise

-0.68

ylon

-0.67

naire

-0.65

volt

-0.65

POSITIVE LOGITS

 sides

1.52

 sexes

1.40

 genders

1.23

 parties

1.23

 halves

1.15

 versions

0.98

 Houses

0.98

 coasts

0.95

 directions

0.90

 organizations

0.89

Activations Density 0.058%

mentions of similarities or commonalities between two entities

instances of the word "Both."

No Comments

No Known Activations