INDEX

Explanations

words related to tables or lists

oai_token-act-pair · gpt-3.5-turbo

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

vernment

-0.83

 Directorate

-0.72

alez

-0.66

rily

-0.65

 Mehran

-0.64

eson

-0.61

 holiest

-0.61

 Rated

-0.61

 Markus

-0.60

adelphia

-0.59

POSITIVE LOGITS

cloth

1.58

au

1.01

aux

1.00

 manners

0.96

tops

0.94

poons

0.93

top

0.93

 scraps

0.86

 tennis

0.85

poon

0.84

Activations Density 0.033%

words related to tables or lists

No Comments

No Known Activations