INDEX

Explanations

segments related to programming code structures and functions

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

Code, properties, or configuration snippets

np_acts-logits-general · gemini-2.0-flash

programming concepts and technical terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

-0.56

<eos>

-0.55

-0.50

//

-0.49

↵↵

-0.49

-0.46

<strong>

-0.44

<h2>

-0.44

-0.42

POSITIVE LOGITS

 GenerationType

1.07

 betweenstory

1.00

脚注の使い方

0.98

transQ

0.96

RegressionTest

0.81

0.80

AddTagHelper

0.80

thâu

0.74

featureID

0.74

berdayakan

0.74

Activations Density 0.116%

segments related to programming code structures and functions

Code, properties, or configuration snippets

programming concepts and technical terms

No Comments

No Known Activations