INDEX

Explanations

numerical references, specifically the number "one" in various contexts

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

references to the concept of unity or singularity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post

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_post

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

grounds

-0.75

scenes

-0.70

lov

-0.68

ãĤ¤ãĥĪ

-0.68

ãĤ¬

-0.66

thumbnails

-0.66

bows

-0.65

videos

-0.65

veyard

-0.65

è¦ļéĨĴ

-0.64

POSITIVE LOGITS

 sided

0.97

 another

0.97

's

0.96

 hundred

0.89

 hemisphere

0.89

 particular

0.88

 side

0.86

 dimensional

0.85

another

0.83

 embodiment

0.80

Activations Density 0.110%

numerical references, specifically the number "one" in various contexts

references to the concept of unity or singularity

No Comments

No Known Activations