INDEX

Explanations

specific names and terms related to people and brands in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

barrow

-1.07

 Mourinho

-1.06

 AssemblyCulture

-1.00

ьаж

-0.97

 Hutchins

-0.90

 Rejo

-0.90

 Chernobyl

-0.90

DPM

-0.89

AFX

-0.88

 Dany

-0.87

POSITIVE LOGITS

 Elis

0.78

 useDispatch

0.66

pp

0.65

boldsymbol

0.65

 Beasts

0.65

LoginPage

0.65

 iprot

0.64

 LAMB

0.64

prima

0.63

geführt

0.63

Activations Density 2.736%

specific names and terms related to people and brands in various contexts

No Comments

No Known Activations