INDEX

Explanations

sections of text that denote significant information or conclusions, possibly related to academic or research findings

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

text-structure and formatting markers (section headings, separators and other layout/markup artifacts).

oai_token-act-pair · gpt-5-mini Triggered by @phamvanluong2003

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 الرياضيه

-0.72

 bezeichneter

-0.68

ècie

-0.67

 /\.(

-0.67

 FishBase

-0.66

 Paglinawan

-0.65

+#+

-0.62

Население

-0.62

енча

-0.61

°•

-0.60

POSITIVE LOGITS

 }}"></

0.65

bbene

0.65

</h6>

0.60

akad

0.56

 ruff

0.53

 eingeb

0.53

 cucchiaio

0.52

DoubleQuotes

0.52

JoinTable

0.52

 biscuit

0.51

Activations Density 0.092%

sections of text that denote significant information or conclusions, possibly related to academic or research findings

text-structure and formatting markers (section headings, separators and other layout/markup artifacts).

No Comments

No Known Activations