INDEX

Explanations

quotations or reported speech

oai_token-act-pair · gpt-3.5-turbo

punctuation marks, particularly those used in dialogue

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

Ĥª

-0.90

ramid

-0.79

artifacts

-0.73

rats

-0.73

exempt

-0.72

ļéĨĴ

-0.71

opian

-0.71

PP

-0.71

¬¼

-0.70

ĻĤ

-0.70

POSITIVE LOGITS

 exclaimed

1.24

 replied

1.22

 muttered

1.21

 whispered

1.20

 says

1.13

 shouted

1.13

 reads

1.12

 yelled

1.09

 cried

1.08

 remarked

1.08

Activations Density 0.112%

quotations or reported speech

punctuation marks, particularly those used in dialogue

No Comments

No Known Activations