INDEX

Explanations

references to communication events and documentation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ema

-0.17

-0.16

eo

-0.16

ems

-0.16

zb

-0.16

-0.15

-0.14

kl

-0.14

dpi

-0.14

POSITIVE LOGITS

KeyType

0.17

AsyncResult

0.16

riend

0.15

ï»¿using

0.15

inet

0.15

 putas

0.15

LinkId

0.15

orders

0.15

MainThread

0.14

 Bris

0.14

Activations Density 0.098%

references to communication events and documentation

No Comments

No Known Activations