INDEX

Explanations

phrases related to accusations and testimonies in a legal or investigative context

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

accusations and claims

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

argout

-0.51

iettivo

-0.50

сылкі

-0.48

textTheme

-0.48

səhifə

-0.45

 aanges

-0.45

Mako

-0.44

rceil

-0.44

stateParams

-0.44

 ImportError

-0.43

POSITIVE LOGITS

 revelations

0.51

writeFieldEnd

0.45

msgTypes

0.44

 révé

0.42

PostInfinity

0.42

tonode

0.39

 chứng

0.36

 revealing

0.36

 reveal

0.36

FormTagHelper

0.36

Activations Density 0.044%

phrases related to accusations and testimonies in a legal or investigative context

accusations and claims

No Comments

No Known Activations