INDEX

Explanations

sections related to operational details and references

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

section labels and headers

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

MigrationBuilder

-0.78

+#+#

-0.76

 հղումներ

-0.67

UrlResolution

-0.66

 doInBackground

-0.66

 препратки

-0.64

UnusedPrivate

-0.62

enderror

-0.60

ImageContext

-0.59

 оригіналу

-0.59

POSITIVE LOGITS

 hours

0.48

Hours

0.44

 Hours

0.42

hours

0.40

antry

0.33

 publick

0.33

 kort

0.32

 duro

0.32

 requirements

0.32

 diting

0.31

Activations Density 0.001%

sections related to operational details and references

section labels and headers

No Comments

No Known Activations