INDEX

Explanations

references to total counts or total quantities in a context, particularly regarding data or results

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

setVerticalGroup

-0.51

principalColumn

-0.48

 vícti

-0.45

хьтан

-0.44

 recrutement

-0.43

 CURIAM

-0.43

 Mexique

-0.42

 cassettes

-0.42

 astore

-0.41

eira

-0.41

POSITIVE LOGITS

 total

3.08

total

2.61

 Total

2.27

 TOTAL

2.22

Total

2.19

 totale

2.06

TOTAL

2.05

 totales

1.82

 totaal

1.76

 totalt

1.70

Activations Density 0.201%

references to total counts or total quantities in a context, particularly regarding data or results

No Comments

No Known Activations