INDEX

Explanations

expressions indicating the impact on individuals or groups in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 41-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_41/width_16k/average_l0_113

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.41.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



-0.49

CppCodeGen

-0.48

 ligiloj

-0.47

posedge

-0.45

Mostly

-0.42

essentiel

-0.42

三年

-0.42

SerializedSize

-0.41

Months

-0.40

三天

-0.40

POSITIVE LOGITS

 many

3.06

many

2.19

 muchos

1.91

Many

1.90

 många

1.85

 vielen

1.84

 viele

1.84

 muitos

1.82

 Many

1.81

 многих

1.72

Activations Density 0.398%

expressions indicating the impact on individuals or groups in various contexts

No Comments

No Known Activations