INDEX

Explanations

structures or elements related to templates and function definitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

and

-0.52

andte

-0.50

-0.49

-0.47

-0.46

ven

-0.45

 yüzden

-0.45

-0.44

The

-0.44

 alá

-0.44

POSITIVE LOGITS

setVerticalGroup

1.19

fjspx

1.05

abinieri

0.93

 مرئيه

0.90

});*/

0.89

kháu

0.88

 nakalista

0.87

rungsseite

0.87

 propOrder

0.87

verwijspagina

0.86

Activations Density 0.030%

structures or elements related to templates and function definitions

No Comments

No Known Activations

structures or elements related to templates and function definitions

No Comments

No Known Activations