INDEX

Explanations

instances of examples or illustrative cases in the text

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

"example" or "examples"

np_acts-logits-general · gemini-2.0-flash

example code or explanations

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 ſtate

-0.44

 pleaſure

-0.42

 ſta

-0.42

 uſe

-0.42

 ſtand

-0.40

 purpoſe

-0.39

Hochspringen

-0.38

 dieux

-0.38

 faſt

-0.37

 pouvoirs

-0.37

POSITIVE LOGITS

 examples

1.09

 example

0.97

 Examples

0.90

Examples

0.90

 Beispiel

0.89

 Beispiele

0.89

 Example

0.88

 exemplos

0.86

 ejemplos

0.85

 exemplo

0.85

Activations Density 0.590%

instances of examples or illustrative cases in the text

"example" or "examples"

example code or explanations

No Comments

No Known Activations