INDEX

Explanations

references to people and their actions or feelings, particularly in a conversational context

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 increí

-0.63

OGND

-0.61

исленность

-0.61

ⓧ

-0.56

bootstrapcdn

-0.55

 autorytatywna

-0.53

IVEREF

-0.52

 ſeine

-0.52

 indígen

-0.52

 Nationen

-0.51

POSITIVE LOGITS

 guys

0.40

 Lordships

0.36

 knob

0.35

propTypes

0.32

 kindly

0.32

 молод

0.31

 char

0.30

う

0.30

van

0.29

IllegalArgument

0.29

Activations Density 0.188%

references to people and their actions or feelings, particularly in a conversational context

No Known Activations