INDEX

Explanations

technical terms and references related to programming languages and platforms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ei

-0.16

rink

-0.15

ÑĢÐ¾Ñĩ

-0.14

plex

-0.14

lez

-0.14

ëģ¼

-0.13

 ringing

-0.13

 ring

-0.13

zed

-0.13

obile

-0.13

POSITIVE LOGITS

uja

0.16

ÐµÐ²Ð¸Ñĩ

0.15

ueil

0.15

UPER

0.14

ayet

0.14

å¤§åħ¨

0.14

 bordel

0.14

ehler

0.14

Mahon

0.14

Lbl

0.13

Activations Density 0.015%

technical terms and references related to programming languages and platforms

No Comments

No Known Activations