INDEX

Explanations

statements that assert factual information or strong claims

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-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

بوابة

-0.79

Portail

-0.54

>>>>>>>

-0.53

 חיצוניים

-0.49

 ARISING

-0.49

ніципалі

-0.47

 colorWith

-0.47

寨

-0.47

 continuant

-0.46

 transportasi

-0.46

POSITIVE LOGITS

 entanto

0.97

etheless

0.88

anmoins

0.78

 belonged

0.74

 belong

0.74

 belongs

0.70

 FetchType

0.68

 intact

0.68

án

0.67

Viitteet

0.66

Activations Density 0.067%

statements that assert factual information or strong claims

No Comments

No Known Activations