INDEX

Explanations

references to organizational affiliations or sources

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 fVar

-0.73

 tunnels

-0.68

чина

-0.63

 abstra

-0.61

––––

-0.58

pezial

-0.58

 Fide

-0.58

Kob

-0.58

 Tully

-0.56

quill

-0.56

POSITIVE LOGITS

org

1.90

ORG

1.66

ORG

1.23

Org

1.19

Org

1.00

orgi

0.90

bootstrapcdn

0.88

org

0.87

orgs

0.86

 Jorg

0.83

Activations Density 0.065%

references to organizational affiliations or sources

No Comments

No Known Activations