INDEX

Explanations

phrases where someone is expressing an opinion or judgment

oai_token-act-pair · gpt-3.5-turbo

descriptive language and terms relating to characterization and evaluation

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 sqor

-0.56

existing

-0.55

ADS

-0.53

DAQ

-0.52

dinand

-0.52

?'

-0.52

asaki

-0.47

MAS

-0.47

ablishment

-0.47

restling

-0.46

POSITIVE LOGITS

"'

2.19

 "âĢ¦

2.11

 "...

2.06

"[

2.04

1.96

"#

1.78

".

1.73

"(

1.72

"@

1.59

",

1.56

Activations Density 0.755%

phrases where someone is expressing an opinion or judgment

descriptive language and terms relating to characterization and evaluation

No Comments

No Known Activations