INDEX

Explanations

references to web-related topics and functionalities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

Efq

-0.85

 Chriftian

-0.77

 Theſe

-0.77

 myſelf

-0.76

 themſelves

-0.75

 Monfieur

-0.73

 againſt

-0.71

 Shakspeare

-0.71

 ſeveral

-0.71

 ſhe

-0.70

POSITIVE LOGITS

 propOrder

0.68

0.60

makeConstraints

0.60

esModule

0.59

dAtA

0.59

 AssemblyCulture

0.58

ftagPool

0.55

pan

0.55

ngdoc

0.55

du

0.54

Activations Density 0.331%

references to web-related topics and functionalities

No Comments

No Known Activations