INDEX

Explanations

tokens that indicate values or references involving data structure syntax and formatting

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

addCriterion

-0.75

^(@)

-0.68

 TimeUnit

-0.65

PropertyChanging

-0.63

AsUp

-0.63

 createContext

-0.56

-0.54

 Jefus

-0.54

-0.53

 mergeFrom

-0.53

POSITIVE LOGITS

AddTagHelper

0.71

 مشين

0.60

 desta

0.54

COUVER

0.53

gypti

0.53

<bos>

0.52

<eos>

0.52

doise

0.51

 推荐

0.51

}}}

0.50

Activations Density 1.208%

tokens that indicate values or references involving data structure syntax and formatting

No Comments

No Known Activations