INDEX

Explanations

recommendations for improvement or optimal choices in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

duk

-0.17

zeich

-0.17

ieri

-0.15

mdp

-0.15

Î³Î¬

-0.14

mere

-0.14

á»ī

-0.14

otec

-0.14

ither

-0.14

ispers

-0.14

POSITIVE LOGITS

 served

0.20

 bets

0.18

-su

0.18

 sticking

0.17

 option

0.17

 suited

0.16

idge

0.16

/e

0.16

err

0.16

wise

0.16

Activations Density 0.062%

recommendations for improvement or optimal choices in various contexts

No Comments

No Known Activations