INDEX

Explanations

specific patterns in numerical data or code-like structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

BeginContext

-0.57

Label

-0.56

 labels

-0.55

::$

-0.54

MENAFN

-0.53

principalTable

-0.52

 Labels

-0.50

 Label

-0.50

出版年

-0.49

label

-0.49

POSITIVE LOGITS

 AppComponent

0.63

StoryboardSegue

0.59

IsMutable

0.59

AsUp

0.58

 viewType

0.56

stdc

0.56

AddTagHelper

0.56

 BoxDecoration

0.56

 EClass

0.53

SourceChecksum

0.53

Activations Density 0.374%

specific patterns in numerical data or code-like structures

No Comments

No Known Activations