INDEX

Explanations

phrases indicating continuation or progression of a list or series

oai_token-act-pair · gpt-3.5-turbo

repetitive phrases that indicate continuation or elaboration on a subject

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

Kids

-0.69

 Kids

-0.63

liter

-0.61

NF

-0.60

Fight

-0.60

 Cree

-0.59

 Dude

-0.57

notice

-0.57

 Purg

-0.56

WOR

-0.56

POSITIVE LOGITS

 forth

1.68

forth

1.17

othe

1.06

apy

0.96

far

0.94

oths

0.90

bered

0.86

ooo

0.85

oner

0.85

othes

0.85

Activations Density 0.045%

phrases indicating continuation or progression of a list or series

repetitive phrases that indicate continuation or elaboration on a subject

No Comments

No Known Activations