INDEX

Explanations

programming language constructs and code-related terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

__((

-0.54

Tikang

-0.54

 nakalista

-0.49

Rüyada

-0.44

RTEE

-0.43

edenken

-0.43

ancey

-0.43

SpringBootTest

-0.42

BibitemShut

-0.42

DataSnapshot

-0.42

POSITIVE LOGITS

 esternos

0.54

 ſhall

0.45

 ſhould

0.45

 Monfieur

0.43

 ſaid

0.40

 alſo

0.40

 theſe

0.39

/**

0.39

 defaultstate

0.38

 obſ

0.38

Activations Density 0.374%

programming language constructs and code-related terminology

No Comments

No Known Activations