INDEX

Explanations

terms related to publications or issues, particularly in a series or collection

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

issue / issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 Siempre

-0.35

Хьажоргаш

-0.34

Later

-0.33

 later

-0.31

 Winaray

-0.31

<?

-0.29

later

-0.29

 Exteriores

-0.29

Toujours

-0.28

SharedDtor

-0.28

POSITIVE LOGITS

 Issue

2.58

Issue

2.41

 ISSUE

2.02

issue

1.88

ISSUE

1.84

 issue

1.77

 Issues

1.52

Issues

1.40

issues

1.35

 issues

1.29

Activations Density 0.004%

terms related to publications or issues, particularly in a series or collection

issue / issues

No Comments

No Known Activations