INDEX

Explanations

programming-related syntax or structures, especially in the context of code generation and placeholder methods

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_11/width_16k/average_l0_118

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

 referrerpolicy

-0.60

multirow

-0.50

 Exacts

-0.47

ceras

-0.44

elas

-0.41

resser

-0.40

eseorang

-0.39

 qualified

-0.39

нской

-0.39

هما

-0.38

POSITIVE LOGITS

 BorderSide

0.61

}{*}{

0.59

|}{

0.56

بوابة

0.55

__":

0.55

|}{$

0.54

 "..\..\

0.52

__':

0.52

featureID

0.52

Portail

0.52

Activations Density 0.234%

programming-related syntax or structures, especially in the context of code generation and placeholder methods

No Comments

No Known Activations