INDEX

Explanations

references to research findings or results

oai_token-act-pair · gpt-4o-mini

New Auto-Interp

Configuration

chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.12.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Architecture

jumprelu

Context Size

1,024

Dataset

chanind/pile-uncopyrighted-gemma-1024-abbrv-1B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Roberts

-0.77

Roberts

-0.60

 control

-0.55

(!__

-0.54

 estimés

-0.54

 المعيارى

-0.54

オク

-0.53

Me

-0.53

 trông

-0.53

Ne

-0.51

POSITIVE LOGITS

 Findings

1.00

 FINDINGS

0.92

Findings

0.92

 Majefty

0.91

 facie

0.91

 doInBackground

0.87

 Theſe

0.86

Autoritní

0.85

 findings

0.85

 dataSnapshot

0.83

Activations Density 0.016%

references to research findings or results

No Comments

No Known Activations