INDEX

Explanations

words related to actions that involve interaction or communication with others

oai_token-act-pair · gpt-3.5-turbo

occurrences of the article "a"

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

aim

-0.63

 Encyclopedia

-0.63

 insult

-0.63

LSD

-0.60

 Illum

-0.59

 excuse

-0.58

ECB

-0.58

easy

-0.58

},{"

-0.56

 opium

-0.56

POSITIVE LOGITS

ria

0.78

 guiActiveUn

0.78

plin

0.74

riel

0.74

rial

0.74

lex

0.73

aron

0.71

UTH

0.70

ilee

0.69

UG

0.69

Activations Density 0.146%

words related to actions that involve interaction or communication with others

occurrences of the article "a"

No Comments

No Known Activations