INDEX

Explanations

keywords related to groups, organizations, or institutions

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

concepts and categories

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

TheReal

-0.54

]")]

-0.53

^(@)

-0.53

 />";

-0.45

MIDDLEWARE

-0.45

 />';

-0.44

HasForeignKey

-0.44

IAm

-0.44

CGRectMake

-0.43

 bParam

-0.43

POSITIVE LOGITS

 resourceCulture

0.55

webtoken

0.52

uxxxx

0.49

 rodilla

0.47

 nutrición

0.46

 bocetos

0.45

 cocaína

0.45

 dewasa

0.45

 météo

0.44

ähteet

0.44

Activations Density 0.365%

keywords related to groups, organizations, or institutions

concepts and categories

No Comments

No Known Activations