INDEX

Explanations

comparative phrases expressing a sense of degree or difference

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 34-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_34/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.34.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

tagHelperRunner

-0.43

src

-0.41

//

-0.36

 situation

-0.36

postIndex

-0.35

 pied

-0.34

ArgsConstructor

-0.33

[+]

-0.33

 meanwhile

-0.33

(.)

-0.32

POSITIVE LOGITS

 usual

0.86

usual

0.79

 zwykle

0.70

kyou

0.66

 ніж

0.66

necessary

0.65

 ever

0.65

 prévu

0.65

 THAN

0.65

 mitä

0.65

Activations Density 0.153%

comparative phrases expressing a sense of degree or difference

No Comments

No Known Activations