INDEX

Explanations

terms related to healing and repair processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/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.21.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

getElements

-0.47

 itemView

-0.47

 Dominus

-0.46

Vee

-0.43

XAxis

-0.42

hasMoreElements

-0.41

setLayoutManager

-0.41

Daniel

-0.40

 Daniel

-0.40

ftime

-0.39

POSITIVE LOGITS

Repair

1.02

 repair

1.02

repair

1.01

 reparación

1.00

 Repair

0.98

 repairs

0.94

 recovery

0.93

 reparation

0.93

 repairing

0.92

 REPAIR

0.92

Activations Density 0.430%

terms related to healing and repair processes

No Comments

No Known Activations