INDEX

Explanations

dialogue punctuation and communication cues like "said," "?," and "!" in conversations

oai_token-act-pair · gpt-3.5-turbo

punctuation marks that signify dialogue and emotional expressions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

eering

-0.75

Thumbnail

-0.73

Ĥª

-0.72

displayText

-0.67

respective

-0.67

PHOTOS

-0.66

anship

-0.66

agine

-0.66

conservancy

-0.66

oreal

-0.65

POSITIVE LOGITS

 replied

1.14

 muttered

1.12

 exclaimed

1.10

 whispered

1.06

she

1.05

 cried

1.02

 shouted

1.02

 yelled

0.98

 joked

0.96

he

0.95

Activations Density 0.099%

dialogue punctuation and communication cues like "said," "?," and "!" in conversations

punctuation marks that signify dialogue and emotional expressions

No Comments

No Known Activations