INDEX

Explanations

parts of code related to class and method definitions

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

Code snippets with parentheses and code-related terms

np_acts-logits-general · gemini-2.0-flash

closing parentheses )

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

Autoritní

-0.73

 ſche

-0.71

RetentionPolicy

-0.68

 paſſ

-0.66

 ſte

-0.66

 ſon

-0.64

 deſt

-0.64

 ſtre

-0.64

 houſe

-0.64

 pleaſure

-0.63

POSITIVE LOGITS

=>

0.76

=>

0.59

 kjer

0.46

 then

0.44

=>

0.41

=>$

0.39

 என்ற

0.39

↵

0.39

 this

0.39

lieger

0.38

Activations Density 0.036%

parts of code related to class and method definitions

Code snippets with parentheses and code-related terms

closing parentheses )

No Comments

No Known Activations