INDEX

Explanations

technical terms and software-related content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

 questions

-0.21

 question

-0.17

McM

-0.17

 Questions

-0.16

AWN

-0.15

 shared

-0.15

 exporter

-0.15

 savoir

-0.15

 Question

-0.14

agic

-0.14

POSITIVE LOGITS

 íĦ°

0.17

rtl

0.16

ohl

0.16

æīĵ

0.16

 addCriterion

0.15

 barcode

0.15

 Speedway

0.14

 ÚĨØ§Ù¾

0.14

 rugged

0.14

echa

0.14

Activations Density 0.027%

technical terms and software-related content

No Comments

No Known Activations