INDEX

Explanations

numerical or code-like information, possibly related to military or aviation contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_33/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.33.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

 informée

-0.58

-0.57

RegressionTest

-0.55

stateProvider

-0.54

rungsseite

-0.53

 transfieras

-0.52

URLException

-0.51

ValueGenerated

-0.50

 Picchu

-0.50

Cyfeiriadau

-0.49

POSITIVE LOGITS

 fighter

0.91

 Fighter

0.79

Fighter

0.68

fighter

0.65

 fighters

0.63

 bomber

0.63

MiG

0.62

 pilot

0.61

MIG

0.59

 pilots

0.58

Activations Density 0.254%

numerical or code-like information, possibly related to military or aviation contexts

No Comments

No Known Activations