INDEX

Explanations

references to ordinary human experiences and characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

مصادر

-0.48

ytä

-0.47

 Goed

-0.45

StrictEqual

-0.45

 nakalista

-0.45

taines

-0.45

晦

-0.44

奉

-0.43

loc

-0.43

-0.42

POSITIVE LOGITS

 humanas

0.89

 ordinary

0.85

 humaine

0.78

 humain

0.78

 AssemblyCulture

0.77

ordinary

0.76

GEBURTSDATUM

0.76

 humano

0.74

istoitu

0.74

 relatable

0.73

Activations Density 0.259%

references to ordinary human experiences and characteristics

No Comments

No Known Activations