INDEX

Explanations

terms related to military and health contexts, particularly involving veterans and medical conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 Begle

-0.48

ագրություններ

-0.46

 Venise

-0.44

 orilla

-0.43

 logam

-0.43

 Eğer

-0.42

 verst

-0.40

 almohada

-0.40

 Theorie

-0.39

alakip

-0.39

POSITIVE LOGITS

AddTagHelper

0.68

WriteTagHelper

0.65

vet

0.64

 slash

0.60

帖最后由

0.59

 Slash

0.59

Mia

0.58

Mia

0.56

Broad

0.56

ened

0.56

Activations Density 0.327%

terms related to military and health contexts, particularly involving veterans and medical conditions

No Comments

No Known Activations