INDEX

Explanations

code-related elements, particularly definitions and types in a programming context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 समीक्षाएं

-0.67

 kasarigan

-0.64

DES

-0.63

-0.62

Des

-0.61

Des

-0.60

afficheront

-0.58

opa

-0.58

DEC

-0.57

let

-0.57

POSITIVE LOGITS

 uſed

0.90

 ainfi

0.84

NAME

0.81

TYPE

0.79

 zelve

0.76

TABLE

0.75

RESULTS

0.73

 auroit

0.73

 préfère

0.73

 doubtnut

0.73

Activations Density 0.257%

code-related elements, particularly definitions and types in a programming context

No Comments

No Known Activations