INDEX

Explanations

chess-related terminology and concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

izo

-0.14

amon

-0.14

ÐºÑģ

-0.13

psilon

-0.13

ayed

-0.13

bk

-0.13

urg

-0.13

istrat

-0.13

.createObject

-0.13

 prere

-0.13

POSITIVE LOGITS

Ã±ana

0.17

 Lives

0.16

uitka

0.16

edla

0.15

 scooter

0.14

iston

0.14

onald

0.14

kea

0.14

0.13

ocha

0.13

Activations Density 0.002%

chess-related terminology and concepts

No Comments

No Known Activations