INDEX

Explanations

identifiers or commands associated with various users or topics in dialogue

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 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

Enllaces

-0.59

Sklici

-0.54

AddTagHelper

-0.53

 AssemblyCulture

-0.53

dealloc

-0.52

 مشين

-0.52

Pyx

-0.51

BufferException

-0.50

 ziekte

-0.49

 Weller

-0.48

POSITIVE LOGITS

AlterField

0.63

 also

0.54

 Catawiki

0.54

肝

0.51

wom

0.51

 rapist

0.49

cheduler

0.49

 上下

0.48

 preferably

0.48

 Routed

0.47

Activations Density 0.307%

identifiers or commands associated with various users or topics in dialogue

No Comments

No Known Activations