INDEX

Explanations

terms associated with academia and classification ratings

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 combineReducers

-0.51

capaian

-0.45

 officio

-0.44

tonsoft

-0.44

 još

-0.44

 afgelopen

-0.42

 parution

-0.42

leggen

-0.42

 moeite

-0.41

 poslední

-0.40

POSITIVE LOGITS

Tikang

0.75

ensement

0.74

AddTagHelper

0.69

IntoConstraints

0.65

 Sucher

0.64

RenderAtEndOf

0.63

 Samaria

0.62

########.

0.62

campal

0.61

cein

0.61

Activations Density 0.206%

terms associated with academia and classification ratings

No Comments

No Known Activations