INDEX

Explanations

references to programming contexts and APIs

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

context and application code

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

Tikang

-0.88

})->

-0.77

RenderAtEndOf

-0.70

 Signalez

-0.67

 betweenstory

-0.66

########.

-0.66



-0.65

 OMITBAD

-0.65

 سكانية

-0.63

:✨

-0.61

POSITIVE LOGITS

Context

0.56

 mennesker

0.44

 cucchiai

0.44

getApplication

0.43

App

0.40

Application

0.40

ExecuteNonQuery

0.40

 Context

0.40

ExecuteReader

0.39

 Application

0.39

Activations Density 0.000%

references to programming contexts and APIs

context and application code

No Comments

No Known Activations