INDEX

Explanations

the mention of specific numerical values

oai_token-act-pair · gpt-3.5-turbo

quantifiable data and metrics related to pages, sizes, and counts

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

eston

-0.61

nce

-0.61

IELD

-0.60

ocument

-0.59

aina

-0.57

roo

-0.57

 Cloak

-0.56

ror

-0.55

AMI

-0.55

Fol

-0.55

POSITIVE LOGITS

 apiece

1.81

poons

1.48

per

1.22

 total

1.18

 consecut

1.14

 totaling

1.13

max

0.99

 maximum

0.96

 each

0.93

 crammed

0.93

Activations Density 0.296%

the mention of specific numerical values

quantifiable data and metrics related to pages, sizes, and counts

No Comments

No Known Activations