INDEX

Explanations

comparative structures indicating one thing is better or worse than another

oai_token-act-pair · gpt-3.5-turbo

comparative phrases emphasizing the word "than."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Juda

-0.63

Ire

-0.58

èĢ

-0.55

ãĥ

-0.55

 derog

-0.55

 Ranger

-0.54

 stead

-0.54

 tandem

-0.53

 asset

-0.52

 Central

-0.52

POSITIVE LOGITS

atos

1.03

lihood

0.96

xual

0.89

tz

0.85

gs

0.81

pload

0.80

gins

0.80

ply

0.78

onyms

0.78

acle

0.78

Activations Density 0.028%

comparative structures indicating one thing is better or worse than another

comparative phrases emphasizing the word "than."

No Comments

No Known Activations