INDEX

Explanations

code-related terms and structure in programming languages

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

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

 nonetheless

-0.38

swe

-0.34

 nevertheless

-0.34

'&:

-0.32

 encara

-0.32

still

-0.32

awtextra

-0.31

 utan

-0.30

 equally

-0.30

流石

-0.30

POSITIVE LOGITS

+#+#

0.73

########.

0.69

 виправивши

0.65

 AssemblyCulture

0.65

 MainAxisSize

0.63



0.63

AsUp

0.61

transQ

0.60

ThroughAttribute

0.60

 BorderSide

0.59

Activations Density 0.001%

code-related terms and structure in programming languages

No Comments

No Known Activations