INDEX

Explanations

terms related to wound healing and tissue response

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

site or scene

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

EDEFAULT

-0.86

 مشين

-0.67

 transfieras

-0.66

ngdoc

-0.59

acterísticas

-0.59

StructEnd

-0.58

Vidite

-0.57

Viitteet

-0.57

astify

-0.57

 '\\;'

-0.56

POSITIVE LOGITS

 site

1.07

 scene

1.04

 lokasi

0.83

 sites

0.81

scene

0.78

現場

0.78

 escena

0.75

现场

0.74

site

0.73

 location

0.72

Activations Density 0.115%

terms related to wound healing and tissue response

site or scene

No Comments

No Known Activations