INDEX

Explanations

phrases or compounds starting with 'big-'

oai_token-act-pair · gpt-3.5-turbo

references to "big" concepts, particularly involving significant events or notable situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Solitaire

-0.79

pta

-0.72

 Sylvia

-0.71

 Plato

-0.68

 Cambodia

-0.67

 solicitation

-0.67

 Pixie

-0.66

 Powell

-0.66

 Canaver

-0.65

 Scrib

-0.63

POSITIVE LOGITS

scale

1.08

bodied

1.03

eyed

1.03

enough

0.97

dollar

0.95

sized

0.94

capacity

0.92

armed

0.92

ticket

0.90

picture

0.89

Activations Density 0.070%

phrases or compounds starting with 'big-'

references to "big" concepts, particularly involving significant events or notable situations

No Comments

No Known Activations