INDEX

Explanations

terms related to rebuilding and design refinement processes

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

 resourceName

-0.15

anton

-0.15

 Ages

-0.15

 unre

-0.15

acker

-0.14

outh

-0.14

pla

-0.14

aan

-0.14

ìĥĪ

-0.14

owl

-0.13

POSITIVE LOGITS

atÃ³

0.17

/reset

0.17

/update

0.16

 Cary

0.15

tings

0.15

/re

0.15

mbH

0.15

lick

0.15

als

0.15

apons

0.14

Activations Density 0.123%

terms related to rebuilding and design refinement processes

No Comments

No Known Activations