INDEX

Explanations

actions or commands related to creating or instructing something

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

сделать, объяснить, ставить

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

-0.39

 levels

-0.39

$\

-0.37

ريكا

-0.37

 whenever

-0.36

kip

-0.36

 periods

-0.36

{\

-0.35

POSITIVE LOGITS

 queſta

0.82

 propOrder

0.81

 '\\;'

0.79

posedge

0.77

parsedMessage

0.77

 onPostExecute

0.76

 AssemblyTitle

0.73

ſſung

0.72

iſchen

0.70

tagHelperRunner

0.70

Activations Density 0.004%

actions or commands related to creating or instructing something

сделать, объяснить, ставить

No Comments

No Known Activations