INDEX

Explanations

terms related to personal or specific names.

oai_token-act-pair · gpt-3.5-turbo

keywords related to actions and their consequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

imity

-0.46

 lyrics

-0.46

 Newsp

-0.46

natureconservancy

-0.45

rawdownloadcloneembedreportprint

-0.45

 Wikimedia

-0.44

 Shutterstock

-0.44

 Leilan

-0.44

utenberg

-0.43

(%)

-0.43

POSITIVE LOGITS

®

0.66

«

0.65

·

0.65

ī

0.63

Ĳ

0.63

º

0.59

uch

0.56

»

0.56

¹

0.55

Ń

0.55

Activations Density 2.227%

terms related to personal or specific names.

keywords related to actions and their consequences

No Comments

No Known Activations