INDEX

Explanations

quantitative comparisons, specifically ratios and percentages related to medical data or outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_36/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.36.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

 dessutom

-0.44

 Sehingga

-0.39

 utom

-0.37

setlength

-0.36

cerned

-0.36

 bakgrund

-0.35

 astfel

-0.35

 причем

-0.35

 marito

-0.34

 tarko

-0.33

POSITIVE LOGITS

vs

3.45

 versus

3.38

 Versus

2.67

versus

2.61

Vs

2.61

VS

2.44

vs

2.34

Versus

2.08

VS

1.77

 verses

1.75

Activations Density 1.067%

quantitative comparisons, specifically ratios and percentages related to medical data or outcomes

No Comments

No Known Activations