INDEX

Explanations

references to training, education, and resources in various contexts

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

foreign or domain-specific terms

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

featureID

-0.51

########.

-0.44

ArrowToggle

-0.42

期刊论文

-0.40

AsUp

-0.40

AllowUser

-0.38

oneofs

-0.37

stdc

-0.37

HostException

-0.36

DebuggerNonUser

-0.36

POSITIVE LOGITS

 kaarangay

0.48

विक

0.47

Vidite

0.46

 שוליים

0.46

gettyimages

0.46

хьтан

0.46

oa̍t

0.45

 확인함

0.43

ESM

0.42

市镇

0.42

Activations Density 0.044%

references to training, education, and resources in various contexts

foreign or domain-specific terms

No Comments

No Known Activations