INDEX

Explanations

phrases with a strong emotional or emphatic tone

oai_token-act-pair · gpt-3.5-turbo

pronouns and demonstrative words

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

lining

-0.70

 Weston

-0.65

alties

-0.60

 colle

-0.57

AFL

-0.56

 Naples

-0.56

 climates

-0.56

xp

-0.52

 Lavrov

-0.52

 stemming

-0.52

POSITIVE LOGITS

estine

0.83

%"

0.81

cellence

0.80

.""

0.77

`.

0.70

"—

0.70

'"

0.70

ACTED

0.69

]"

0.69

 nomine

0.69

Activations Density 0.177%

phrases with a strong emotional or emphatic tone

pronouns and demonstrative words

No Comments

No Known Activations