INDEX

Explanations

occurrences of the increment operator (++) in code-related contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_1/width_16k/average_l0_105

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.hook_mlp_out

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

++

-1.78

++

-1.75

(++

-1.14

++.

-1.13

++$

-1.09

++,

-1.08

+++

-0.98

(++

-0.94

++++

-0.79

+++

-0.69

POSITIVE LOGITS

ఔ

0.60

 كورة

0.59

 Gemein

0.51

 Klin

0.51

публи

0.50

 disintegration

0.50

NameInMap

0.49

 Hustle

0.49

 canine

0.48

])->

0.48

Activations Density 0.219%

occurrences of the increment operator (++) in code-related contexts

No Comments

No Known Activations