INDEX

Explanations

patterns related to code syntax and function handling

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

Code or configuration snippets

np_acts-logits-general · gemini-2.0-flash

mathematical and code expressions

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

posedge

-0.65

 žá

-0.53

 mentales

-0.52

ēju

-0.51

 bekas

-0.50

 sociaux

-0.49

 féd

-0.49

pyp

-0.49

 pioggia

-0.49

 battre

-0.48

POSITIVE LOGITS

ⓘ

0.69

 <=",

0.67

Viited

0.61

 propOrder

0.60

alyptus

0.56

UnknownFieldSet

0.54

Välislingid

0.53

DECREF

0.53

̍t

0.53

 متعلقه

0.53

Activations Density 0.153%

patterns related to code syntax and function handling

Code or configuration snippets

mathematical and code expressions

No Comments

No Known Activations