INDEX

Explanations

elements related to academic metrics and references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

ldr

-0.17

avir

-0.16

resenter

-0.15

ÐµÑĢÑĤÐ¸

-0.14

atica

-0.14

erea

-0.14

decrypt

-0.14

ÃŃcul

-0.14

YPE

-0.13

nger

-0.13

POSITIVE LOGITS

als

0.16

PTR

0.15

adr

0.15

 principal

0.14

ialog

0.14

atics

0.14

uni

0.14

 instrumental

0.13

alk

0.13

ek

0.13

Activations Density 0.137%

elements related to academic metrics and references

No Comments

No Known Activations