INDEX

Explanations

references to contradictions and hypocrisy in arguments or behaviors

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

 Rodrig

-0.42

 Ember

-0.39

boot

-0.39

 esprits

-0.39

Rcpp

-0.38

 aceton

-0.38

 GenerationType

-0.38

hots

-0.37

 Steck

-0.37

chts

-0.37

POSITIVE LOGITS

awtextra

0.57

SharedDtor

0.52

клопе

0.51

ंदीखरीदारी

0.49

 tartalomajánló

0.48

aarrggbb

0.47

 gyhoeddwyd

0.47

 alami

0.45

PerformLayout

0.45

 intptr

0.45

Activations Density 1.171%

references to contradictions and hypocrisy in arguments or behaviors

No Comments

No Known Activations