INDEX

Explanations

metrics and comparative measures related to achievements and hierarchies

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

awtextra

-0.44

 comte

-0.41

DebuggerNonUser

-0.40

 Putri

-0.40

tagHelperRunner

-0.39

RTEX

-0.39

ніципалі

-0.38

initComponents

-0.38

 vainilla

-0.38

Carbon

-0.38

POSITIVE LOGITS

 Replacing

0.60

Replacing

0.57

 replacing

0.56

 Replaced

0.55

 replacement

0.52

 replaced

0.51

 replace

0.51

 Replacement

0.51

 overtaking

0.50

 swapping

0.50

Activations Density 0.378%

metrics and comparative measures related to achievements and hierarchies

No Comments

No Known Activations