INDEX

Explanations

phrases starting points for a list or enumeration

oai_token-act-pair · gpt-3.5-turbo

lists or enumerations of items or concepts

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

 guiActiveUnfocused

-0.71

ans

-0.68

amphetamine

-0.67

matter

-0.67

 intrusion

-0.66

existent

-0.65

fulness

-0.65

 Dragonbound

-0.63

ibr

-0.63

endor

-0.61

POSITIVE LOGITS

 some

1.20

 excerpts

1.11

 five

1.08

few

1.01

 three

1.01

 seven

0.98

 SOME

0.97

 several

0.97

two

0.95

 four

0.94

Activations Density 0.050%

phrases starting points for a list or enumeration

lists or enumerations of items or concepts

No Comments

No Known Activations