INDEX

Explanations

conjunctions and conditional phrases indicating relationships between ideas

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_11/width_16k/average_l0_118

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

 segíts

-0.44

subsid

-0.40

 Anbieter

-0.40

serious

-0.39

 descal

-0.39

queles

-0.38

ory

-0.36

chossen

-0.36

champion

-0.36

units

-0.36

POSITIVE LOGITS

 autorytatywna

0.54

 transfieras

0.52

 ApiException

0.48

Diwedd

0.48

rungsseite

0.48

@[+][

0.46

Rhestr

0.46

 gainera

0.46

 opérés

0.45

étés

0.43

Activations Density 0.333%

conjunctions and conditional phrases indicating relationships between ideas

No Comments

No Known Activations

conjunctions and conditional phrases indicating relationships between ideas

No Comments

No Known Activations