INDEX

Explanations

references to exceptions or errors in code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

-0.41

carto

-0.39

"./

-0.39

campista

-0.38

 ['./

-0.38

="./

-0.37

ガニック

-0.36

Gar

-0.36

="/

-0.36

map

-0.35

POSITIVE LOGITS

Exception

2.08

 Exception

2.05

 exception

1.74

exception

1.62

 exceptions

1.61

 Exceptions

1.57

Exceptions

1.39

EXCEPTION

1.38

 excepción

1.34

exceptions

1.30

Activations Density 0.060%

references to exceptions or errors in code

No Comments

No Known Activations

references to exceptions or errors in code

No Comments

No Known Activations