INDEX

Explanations

numerical values related to measurements or times

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

 NEGLIGENCE

-0.14

ãĥĨãĥ«

-0.14

gom

-0.13

 tÃŃm

-0.13

dat

-0.13

ToFit

-0.13

zimmer

-0.13

rimon

-0.13

 sack

-0.13

POSITIVE LOGITS

IDO

0.16

ervas

0.15

finity

0.15

ï»¿using

0.15

zes

0.15

ppo

0.15

finally

0.14

unden

0.14

errupt

0.14

 flush

0.14

Activations Density 0.022%

numerical values related to measurements or times

No Comments

No Known Activations