INDEX

Explanations

key phrases and potential causes related to failures and problems

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

CTS

-0.17

unya

-0.16

 Ð½Ð¸

-0.15

ara

-0.15

Fd

-0.14

ValueChanged

-0.14

ataires

-0.14

à¤ł

-0.14

Hin

-0.14

kov

-0.14

POSITIVE LOGITS

eniable

0.19

ETA

0.15

 slugg

0.15

celik

0.15

amac

0.15

ÑģÑĤÑĢ

0.15

opsy

0.14

 chá»©

0.14

urse

0.14

hled

0.14

Activations Density 0.112%

key phrases and potential causes related to failures and problems

No Comments

No Known Activations