INDEX

Explanations

statistical language patterns and numerical data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 35-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_35/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.35.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

 also

-0.95

 также

-0.85

 також

-0.84

 inoltre

-0.80

 ayrıca

-0.79

 también

-0.78

Also

-0.78

 همچنین

-0.77

also

-0.76

 également

-0.75

POSITIVE LOGITS

 Namely

1.15

 Firstly

1.06

 nämlich

1.00

Firstly

0.97

 firstly

0.93

 namely

0.91

 bowiem

0.88

まず

0.85

一方面

0.85

namely

0.84

Activations Density 0.520%

statistical language patterns and numerical data

No Comments

No Known Activations