INDEX

Explanations

words starting with 'sw'

oai_token-act-pair · gpt-3.5-turbo

instances of the term "sw" 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 @ 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 marginal

-0.64

 Rubin

-0.64

CST

-0.62

 percentile

-0.60

 frust

-0.59

 Gilbert

-0.59

ilon

-0.59

 Dome

-0.58

 Burnett

-0.57

 relig

-0.57

POSITIVE LOGITS

sw

4.40

SW

2.01

Sw

1.97

swe

1.56

Sw

1.55

sword

1.48

sm

1.37

swer

1.37

sp

1.37

sw

1.35

Activations Density 0.003%

words starting with 'sw'

instances of the term "sw" in various contexts

No Comments

No Known Activations