INDEX

Explanations

mentions of specific football teams and their performances

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 kasarigan

-0.61

发表于

-0.53

brtc

-0.49

 tenuto

-0.45

hwa

-0.44

 auprès

-0.43

logitech

-0.43

TRANSL

-0.42

Melody

-0.42

meln

-0.42

POSITIVE LOGITS

 football

0.64

 Football

0.60

football

0.56

Football

0.55

NFL

0.54

NFL

0.52

 FOOTBALL

0.48

 quarterback

0.47

🏈

0.47

 fotbal

0.44

Activations Density 0.299%

mentions of specific football teams and their performances

No Comments

No Known Activations