INDEX

Explanations

phrases related to environmental or contextual changes

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

Floating point numbers

np_acts-logits-general · gemini-2.0-flash

weeks and numerical values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 שוליים

-0.95

 purpoſe

-0.91

 Diſ

-0.85

 Majefty

-0.83

 ſeveral

-0.82

 pleaſure

-0.81

 ſever

-0.79

ſelves

-0.78

ſelf

-0.78

CodeAttribute

-0.78

POSITIVE LOGITS

audiovisuel

0.53

кипедия

0.47

ère

0.47

jedis

0.45

^^^^^^^^

0.45

antd

0.45

 شوند

0.44

ViewType

0.43

Activations Density 0.417%

phrases related to environmental or contextual changes

Floating point numbers

weeks and numerical values

No Comments

No Known Activations