INDEX

Explanations

information about articles and their sources

oai_token-act-pair · gpt-3.5-turbo

indicators of archival content and publication details

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

CHO

-0.60

 alike

-0.57

 Stim

-0.57

 Requ

-0.55

CBO

-0.54

 sequ

-0.54

MAT

-0.53

 Conclusion

-0.53

isively

-0.52

 Anat

-0.52

POSITIVE LOGITS

ãĤ¨ãĥ«

0.76

interstitial

0.69

etimes

0.67

 pengu

0.67

dating

0.67

entially

0.66

teness

0.64

chron

0.63

 "$:/

0.63

lde

0.62

Activations Density 0.054%

information about articles and their sources

indicators of archival content and publication details

No Comments

No Known Activations