INDEX

Explanations

notable numerical values, specifically those related to significant events or statistics

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

shell

-0.15

enger

-0.14

ãĤĵãģ¨

-0.14

wyn

-0.14

 shell

-0.13

äºĭåĭĻ

-0.13

primitive

-0.13

Shell

-0.13

 prec

-0.12

 perse

-0.12

POSITIVE LOGITS

ellig

0.15

UILT

0.14

èŃ·

0.14

ovÃŃ

0.14

lesia

0.14

exampleInput

0.13

ÙĪÙĤÙģ

0.13

urement

0.13

IDb

0.13

/stdc

0.12

Activations Density 0.071%

notable numerical values, specifically those related to significant events or statistics

No Comments

No Known Activations