INDEX

Explanations

phrases indicating rigorous research and evidence

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

extensive research/investigation

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

Composable

-0.63

quant

-0.54

 PARTIC

-0.53

 funnier

-0.51

 piccolo

-0.50

Personendaten

-0.50

 MIGHT

-0.48

FormBuilder

-0.48

Pancake

-0.48

 viewType

-0.48

POSITIVE LOGITS

 extensive

0.85

 extensively

0.82

 Extensive

0.79

Extensive

0.77

extensive

0.72

 many

0.63

 Many

0.59

 nombreuses

0.58

 numerous

0.58

 numerosos

0.58

Activations Density 0.065%

phrases indicating rigorous research and evidence

extensive research/investigation

No Comments

No Known Activations