INDEX

Explanations

phrases indicating responses or reactions in a conversational context

oai_token-act-pair · gpt-3.5-turbo

instances of laughter or expressions of amusement

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

 established

-0.77

 centrally

-0.75

 manufact

-0.75

 utilities

-0.73

 accrued

-0.71

 govern

-0.71

 Volks

-0.70

 accumulated

-0.70

 nutrient

-0.69

 fortified

-0.68

POSITIVE LOGITS

Laughs

2.07

laughs

1.99

laughter

1.69

laugh

1.68

Pause

1.47

pause

1.41

sic

1.36

âĢ¦)

1.28

...)

1.22

sorry

1.19

Activations Density 0.080%

phrases indicating responses or reactions in a conversational context

instances of laughter or expressions of amusement

No Comments

No Known Activations

phrases indicating responses or reactions in a conversational context

instances of laughter or expressions of amusement

No Comments

No Known Activations