INDEX

Explanations

code-related constructs and operations in programming contexts

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

`<` operator

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

uxxxx

-0.57

 Houſe

-0.53

########.

-0.49

numerusform

-0.48

enterOuterAlt

-0.45

 verbund

-0.45

parsedMessage

-0.44

 ſta

-0.44

 houſe

-0.44

 itſelf

-0.43

POSITIVE LOGITS

 comparison

0.66

 comparisons

0.65

 comparing

0.65

 compare

0.63

compare

0.60

comparing

0.60

 Comparison

0.60

 comparer

0.59

 Comparisons

0.59

compar

0.59

Activations Density 0.009%

code-related constructs and operations in programming contexts

`<` operator

No Comments

No Known Activations