INDEX

Explanations

phrases that express knowledge or awareness

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

dhist

-0.71

qxd

-0.65

TEntity

-0.62

Tig

-0.60

 Linton

-0.59

czak

-0.59

 Larsson

-0.59

mathbf

-0.58

wicklung

-0.58

Tos

-0.57

POSITIVE LOGITS

 know

1.82

know

1.80

Know

1.77

 Know

1.74

 knows

1.65

KNOW

1.64

knows

1.59

 KNOW

1.57

 Knows

1.48

 Knowles

1.47

Activations Density 0.705%

phrases that express knowledge or awareness

No Comments

No Known Activations