INDEX

Explanations

proper nouns or names

oai_token-act-pair · gpt-3.5-turbo

the mention of "Col" followed by numerical values or related identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

REM

-0.75

avorite

-0.69

 slack

-0.67

andowski

-0.66

ucks

-0.65

 fumble

-0.65

swing

-0.64

hift

-0.63

lambda

-0.63

tarian

-0.62

POSITIVE LOGITS

Col

3.72

Col

2.35

col

1.94

col

1.88

COL

1.79

COL

1.70

 Colonel

1.52

Maj

1.50

 Colon

1.46

 Brig

1.32

Activations Density 0.022%

proper nouns or names

the mention of "Col" followed by numerical values or related identifiers

No Comments

No Known Activations