INDEX

Explanations

references to natural elements and entities in the context of religious or philosophical discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_33/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.33.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

tvguidetime

-1.02

脚注の使い方

-0.86

 kasarigan

-0.85

tagHelperRunner

-0.82

ReusableCell

-0.79

<pad>

-0.77

 protoimpl

-0.77

Tikang

-0.77

[@BOS@]

-0.76

<unused23>

-0.76

POSITIVE LOGITS

and

0.45

or

0.43

Pfe

0.40

<strong>

0.39

to

0.38

 Schreiber

0.38

↵↵

0.36

 Morel

0.36

 Lenz

0.35

Activations Density 1.819%

references to natural elements and entities in the context of religious or philosophical discussions

No Comments

No Known Activations