INDEX

Explanations

programming and technical terminologies related to data handling and processing

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

Debugging, logging, tracing, or error messages

np_acts-logits-general · gemini-2.0-flash

log/command keywords

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 28-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_28/width_16k/average_l0_119

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.28.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

 feroit

-0.47

 spørs

-0.46

 pinulongan

-0.46

 AssemblyCulture

-0.44

 natomiast

-0.41

 suaminya

-0.41

 himſelf

-0.40

argout

-0.39

ThroughAttribute

-0.39

POSITIVE LOGITS

ok

0.54

OK

0.53

br

0.49

 okay

0.49

0.48

 default

0.47

 existential

0.47

 makeStyles

0.47

 Vars

0.47

dor

0.47

Activations Density 0.306%

programming and technical terminologies related to data handling and processing

Debugging, logging, tracing, or error messages

log/command keywords

No Comments

No Known Activations