INDEX

Explanations

terms related to precision and quality in various contexts

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

"fine" and its related uses

np_acts-logits-general · gemini-2.0-flash

fine tuning and particles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

yntaxException

-0.61

rrggbb

-0.59

קישורים

-0.52

ApiModel

-0.51

يديا

-0.50

المناصب

-0.50

uxxxx

-0.49

 therefor

-0.48

 kaki

-0.48

ickness

-0.46

POSITIVE LOGITS

fine

1.07

 tuned

1.06

Fine

1.06

 tuning

1.05

 fine

1.04

 grained

1.02

 FINE

1.02

 Fine

1.00

FINE

0.96

 tune

0.95

Activations Density 0.086%

terms related to precision and quality in various contexts

"fine" and its related uses

fine tuning and particles

No Comments

No Known Activations