INDEX

Explanations

specific terms related to evidence and analysis in legal and scientific contexts

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

legal context nouns

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

 initComponents

-0.60

 propOrder

-0.59

setVerticalGroup

-0.59

 nakalista

-0.55

aarrggbb

-0.54

BeginContext

-0.52

findpost

-0.52

saraba

-0.48

 برانيه

-0.48

 ErrIntOverflow

-0.47

POSITIVE LOGITS

Gön

0.40

 really

0.40

 McIntyre

0.40

icrous

0.39

 Farina

0.38

жника

0.37

base

0.37

Читати

0.37

ReadKey

0.37

りが

0.36

Activations Density 0.135%

specific terms related to evidence and analysis in legal and scientific contexts

legal context nouns

No Comments

No Known Activations