INDEX

Explanations

phrases related to experimentation and iterative processes

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

learning from trial and error

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

NameInMap

-0.52

ContentLoaded

-0.46

 ujednoznacz

-0.45

COUVER

-0.42

 cervello

-0.41

 conmigo

-0.40

 vertu

-0.40

quelize

-0.40

ſp

-0.40

ambién

-0.40

POSITIVE LOGITS

 experiments

0.51

SequentialGroup

0.48

原始内容存档于

0.45

 experim

0.45

 NSCoder

0.44

 experimentation

0.44

Experiments

0.40

 utafitiHapana

0.40

esModule

0.39

piram

0.39

Activations Density 0.044%

phrases related to experimentation and iterative processes

learning from trial and error

No Comments

No Known Activations