INDEX

Explanations

words containing the suffix "-ible" with high activation values

oai_token-act-pair · gpt-3.5-turbo

words related to conditions of capability or existence

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

 Adren

-0.65

Ren

-0.63

stress

-0.63

 preferring

-0.62

strip

-0.61

 Wass

-0.59

stri

-0.59

 Admir

-0.58

 oscill

-0.54

 opting

-0.54

POSITIVE LOGITS

ible

4.85

ibly

3.20

ibles

3.15

ibility

3.14

IBLE

2.73

ibilities

2.27

able

1.63

ibl

1.62

iable

1.31

uble

1.30

Activations Density 0.008%

words containing the suffix "-ible" with high activation values

words related to conditions of capability or existence

No Comments

No Known Activations