INDEX

Explanations

associations with important concepts or variables within a scientific context

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

associated with

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

 queſta

-1.22

 للمعارف

-1.09

⟬

-1.06

parsedMessage

-1.05

WebVitals

-1.03

<unused52>

-1.02

MigrationBuilder

-1.02

<unused14>

-1.02

<unused8>

-1.02

[@BOS@]

-1.02

POSITIVE LOGITS

0.45

<eos>

0.45

↵↵

0.45

↵↵↵

0.40

0.37

的

0.37

0.35

0.34

Activations Density 0.002%

associations with important concepts or variables within a scientific context

associated with

No Comments

No Known Activations