INDEX

Explanations

references to records 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 @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

opo

-0.17

 addslashes

-0.16

ÏĮÏĦÎ·

-0.15

Ã¤Ã¤

-0.15

.opensource

-0.14

 Starter

-0.14

gtest

-0.14

ppo

-0.14

æĥħ

-0.14

 Rewards

-0.13

POSITIVE LOGITS

 record

0.53

 records

0.53

 Records

0.43

records

0.42

 Record

0.41

 RECORD

0.40

record

0.39

Records

0.38

-record

0.37

 Guinness

0.36

Activations Density 0.035%

references to records and achievements

No Comments

No Known Activations