INDEX

Explanations

numerical values and scores related to competitive events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

 èĹ

-0.16

IColor

-0.15

orners

-0.14

ãĥĭãĥ¥

-0.14

owie

-0.14

éļĨ

-0.13

 Surre

-0.13

 punch

-0.13

lix

-0.13

Î²Îµ

-0.13

POSITIVE LOGITS

Å¡el

0.17

aret

0.15

ettel

0.14

.backend

0.14

 dann

0.14

ÑĢÐ¾Ð²Ð¾Ð´

0.13

-caret

0.13

DNS

0.13

 Barrett

0.13

Activations Density 0.003%

numerical values and scores related to competitive events

No Comments

No Known Activations