INDEX

Explanations

code-related terms and functions associated with parsing or processing data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

\{\\

-0.45

distan

-0.44

featureID

-0.43

Blockly

-0.41

✢

-0.40

inephrine

-0.40

 timeType

-0.40

 Hardy

-0.39

Protect

-0.38

Hardy

-0.38

POSITIVE LOGITS

 processing

0.49

 BeautifulSoup

0.48

InitVars

0.47

 Processing

0.45

processing

0.43

 PROCESSING

0.43

解析

0.42

 decomposition

0.41

 analytical

0.41

 AssemblyTitle

0.40

Activations Density 0.479%

code-related terms and functions associated with parsing or processing data

No Comments

No Known Activations