INDEX

Explanations

concepts related to ability and potential

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_0/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.0.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

 script

-0.45

ter

-0.44

 Roman

-0.42

Gra

-0.41

 Schen

-0.40

 Christensen

-0.40

 German

-0.40

Tre

-0.40

Cas

-0.40

 driver

-0.39

POSITIVE LOGITS

ſelves

0.85

 feroit

0.77

 houſe

0.77

 auffi

0.76

 Anſ

0.75

ſelf

0.74

 auroit

0.74

Want

0.74

 Houſe

0.73

 ſich

0.71

Activations Density 0.308%

concepts related to ability and potential

No Comments

No Known Activations