INDEX

Explanations

expected values in test cases or assertions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

omal

-0.17

lector

-0.16

als

-0.15

benh

-0.15

 past

-0.15

elop

-0.14

alink

-0.14

lew

-0.14

annes

-0.14

aho

-0.14

POSITIVE LOGITS

´

0.15

 ÙĨÙģ

0.14

edom

0.14

sehen

0.14

 ÐĶÐ¸Ð²

0.14

 Luna

0.13

æī¶

0.13

oucher

0.13

lÄ±

0.13

Activations Density 0.020%

expected values in test cases or assertions

No Comments

No Known Activations