INDEX

Explanations

references to survivors and survivor experiences

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

survivor of hardship

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

-0.45

‘

-0.44

-0.42

公

-0.41

 false

-0.40

 fine

-0.40

 English

-0.39

“

-0.39

-0.38

POSITIVE LOGITS

 survivor

1.23

 Survivor

1.17

Survivor

1.16

 Survivors

1.14

 survivors

1.08

Survivors

1.04

parsedMessage

0.96

survi

0.79

 Responder

0.73

 responder

0.72

Activations Density 0.002%

references to survivors and survivor experiences

survivor of hardship

No Comments

No Known Activations