INDEX

Explanations

details about product announcements and specifications

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

product announcements and releases

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

 Seguridad

-0.35

ABETH

-0.35

 Segurança

-0.35

 CRES

-0.35

 podjela

-0.34

ewe

-0.32

 hunger

-0.32

 jang

-0.31

 hoeft

-0.31

 земе

-0.31

POSITIVE LOGITS

 AssemblyVersion

0.68

ArgsConstructor

0.61

 createSlice

0.57

InitVars

0.56

+#+

0.54

 насељу

0.53

 anunciou

0.50

 typelib

0.49

 thiệu

0.49

⎨

0.49

Activations Density 0.014%

details about product announcements and specifications

product announcements and releases

No Comments

No Known Activations