INDEX

Explanations

terms related to official documentation or legal records

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

ider

-0.15

\OptionsResolver

-0.15

nid

-0.15

 KÄ±z

-0.15

uxe

-0.14

aben

-0.13

Äĩi

-0.13

 repmat

-0.13

chap

-0.13

èĵ

-0.13

POSITIVE LOGITS

aday

0.15

reau

0.15

ollo

0.15

QueryBuilder

0.15

ullo

0.15

amina

0.14

RITE

0.14

orne

0.14

moth

0.14

 tall

0.14

Activations Density 0.000%

terms related to official documentation or legal records

No Comments

No Known Activations