INDEX

Explanations

concepts related to methodologies, strategies, and factors in various contexts

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

cases proposals factors theories

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

-0.54

 instead

-0.47

-0.43

another

-0.42

-0.40

-0.38

 without

-0.37

 another

-0.36

 Another

-0.36

POSITIVE LOGITS

 surla

0.81

 propOrder

0.72

 imagui

0.70

 Dieſe

0.67

 ſei

0.67

ロウィン

0.67

 समीक्षाओं

0.67

 パンチラ

0.66

<pad>

0.66

<unused68>

0.66

Activations Density 0.161%

concepts related to methodologies, strategies, and factors in various contexts

cases proposals factors theories

No Comments

No Known Activations