INDEX

Explanations

references to academic competitions and achievements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

-0.15

 conting

-0.15

 Gest

-0.15

 welcome

-0.14

cor

-0.14

imed

-0.14

icus

-0.14

ihat

-0.14

 teach

-0.14

kont

-0.14

POSITIVE LOGITS

asco

0.17

osl

0.17

LOSS

0.14

 çľģ

0.14

-FIRST

0.14

adians

0.14

'gc

0.14

Ì£

0.14

_mD

0.14

WER

0.14

Activations Density 0.040%

references to academic competitions and achievements

No Comments

No Known Activations