INDEX

Explanations

phrases indicating quantity or count

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Millions

-0.18

inski

-0.18

 millions

-0.16

Two

-0.16

two

-0.15

 billions

-0.15

xCD

-0.14

sted

-0.14

 Huge

-0.14

 Much

-0.14

POSITIVE LOGITS

 dozen

0.63

 dozens

0.47

 scores

0.45

 score

0.44

scores

0.37

doz

0.37

 Scores

0.37

 Score

0.36

 handful

0.35

-score

0.35

Activations Density 0.111%

phrases indicating quantity or count

No Comments

No Known Activations