INDEX

Explanations

programming and logging-related commands or functions

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

code snippets and variable names

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

 autorytatywna

-0.71

AddTagHelper

-0.60

 betweenstory

-0.59

uxxxx

-0.59

SequentialGroup

-0.59

SharedDtor

-0.58

berdayakan

-0.56

舺

-0.56

 zoude

-0.56

haikusbot

-0.56

POSITIVE LOGITS

Bakgrunnsstoff

0.40

 אות

0.39

the

0.38

him

0.38

 пи

0.38

raw

0.37

 hire

0.37

 Branche

0.37

 hired

0.37

 fort

0.36

Activations Density 0.006%

programming and logging-related commands or functions

code snippets and variable names

No Comments

No Known Activations