INDEX

Explanations

references to individuals and their actions or existence within a context

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

X が

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

 ActionBar

-0.42

 løpet

-0.41

ക

-0.40

 Kerja

-0.40

 Stirn

-0.40

ing

-0.40

 dieux

-0.38

GTP

-0.36

IfNeeded

-0.35

 لر

-0.35

POSITIVE LOGITS

RegressionTest

0.79

目が

0.76

雰囲気が

0.74

性が

0.73

味が

0.71

時間が

0.69

ьаж

0.68

りが

0.67

物が

0.65

声が

0.65

Activations Density 0.014%

references to individuals and their actions or existence within a context

X が

No Comments

No Known Activations