INDEX

Explanations

references to specific identifiers or classifications in text data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

 Moreno

-0.17

(æ°´

-0.16

Sto

-0.16

anship

-0.15

sto

-0.15

 superf

-0.14

uum

-0.14

embed

-0.14

STORE

-0.14

atz

-0.13

POSITIVE LOGITS

ocht

0.18

Ð»Ð»

0.13

 Muse

0.13

hua

0.13

ola

0.13

 doll

0.13

Sug

0.13

ëĭĪìĬ¤

0.13

 Berm

0.13

erli

0.13

Activations Density 0.028%

references to specific identifiers or classifications in text data

No Comments

No Known Activations