INDEX

Explanations

specific references, numbers, and categorizations within the text

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

 AssemblyCulture

-0.61

Copia

-0.60

ttps

-0.60

 okhttp

-0.54

 pegs

-0.53

fastjson

-0.53

ourt

-0.53

doPost

-0.53

 interferon

-0.53

serialization

-0.52

POSITIVE LOGITS

oredCriteria

0.85

UnusedPrivate

0.80

 respectivamente

0.72

 وتسجيلات

0.64

NameInMap

0.64

etc

0.57

seamnă

0.57

 respectively

0.57

estacks

0.56

RegressionTest

0.56

Activations Density 1.959%

specific references, numbers, and categorizations within the text

No Comments

No Known Activations