INDEX

Explanations

the word "all" in various contexts

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

Архівовано

-0.98

 nahilalakip

-0.94

ীয়

-0.81

)__

-0.80

 greateſt

-0.80

 Luton

-0.80

 publicKey

-0.79

-0.77

 againſt

-0.76

 leaſt

-0.76

POSITIVE LOGITS

All

0.81

MLLoader

0.71

All

0.69

ALL

0.68

all

0.65

 Ganz

0.63

off

0.60

 Dall

0.60

 GetAll

0.60

 just

0.59

Activations Density 0.099%

the word "all" in various contexts

No Comments

No Known Activations