INDEX

Explanations

specific programming constructs and data types related to data structures and manipulation in a coding context

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

abbreviations forming technical words

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

-0.48

ット

-0.42

-0.41

-0.40

-0.38

import

-0.38

 […]

-0.38

Đây

-0.37

<strong>

-0.37

POSITIVE LOGITS

pecabe

0.90

ientras

0.88

ésult

0.85

iſten

0.82

majánló

0.82

ロウィン

0.82

ambién

0.81

<unused43>

0.81

<unused41>

0.81

<unused8>

0.81

Activations Density 0.096%

specific programming constructs and data types related to data structures and manipulation in a coding context

abbreviations forming technical words

No Comments

No Known Activations