INDEX

Explanations

words and phrases related to competitive events or rankings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

ÐµÐ³Ð¾ÑĢ

-0.15

uros

-0.15

eniz

-0.14

ivor

-0.14

ynos

-0.14

initializer

-0.14

auty

-0.14

inel

-0.14

emouth

-0.14

_simps

-0.13

POSITIVE LOGITS

ily

0.15

ally

0.15

Ł

0.15

ãĥĬãĥ¼

0.14

 Garrison

0.14

ops

0.14

Ø¨ÙĬØ±

0.14

 ÃĿ

0.14

alf

0.13

 Democr

0.13

Activations Density 0.010%

words and phrases related to competitive events or rankings

No Comments

No Known Activations