INDEX

Explanations

references to expressions and function calls in a programming context

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

programming expressions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

“

-0.61

 Baumann

-0.57

 Schie

-0.57

 Roberts

-0.56

Initializable

-0.54

“

-0.53

maintenance

-0.53

manage

-0.53

outside

-0.52

 McLeod

-0.52

POSITIVE LOGITS

expr

1.94

 expr

1.88

 Expr

1.70

Expr

1.67

EXPR

1.27

 expre

0.94

 expressions

0.82

texpr

0.79

 Expt

0.78

expre

0.75

Activations Density 0.002%

references to expressions and function calls in a programming context

programming expressions

No Comments

No Known Activations