INDEX

Explanations

nouns and proper nouns associated with individuals and formal settings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

ãĥ³ãĥĦ

-0.16

 Barth

-0.15

ocl

-0.14

frey

-0.14

Ð¾ÑĤÑĮ

-0.14

 Imper

-0.14

hta

-0.14

Î¼Î¼

-0.14

iki

-0.13

asan

-0.13

POSITIVE LOGITS

anova

0.19

lose

0.17

ardi

0.16

ASM

0.16

Î´Î¿

0.16

LOSE

0.15

çļ®

0.15

Ã©p

0.14

à¸§à¸ĩ

0.14

aska

0.14

Activations Density 0.035%

nouns and proper nouns associated with individuals and formal settings

No Comments

No Known Activations