INDEX

Explanations

references to specific brands or figures related to historic and cultural significance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

-0.16

urger

-0.16

_identity

-0.15

à¹Ģà¸Ħà¸£à¸²à¸°à¸«

-0.15

 States

-0.15

áº£y

-0.15

FAULT

-0.14

riter

-0.14

assen

-0.14

POSITIVE LOGITS

acci

0.20

 rack

0.16

 ÐºÑĥÑĢÑģ

0.16

Ore

0.15

aux

0.15

Ð°ÑģÑĤ

0.15

 Reform

0.15

ãĥĪãĥ«

0.14

enor

0.14

iggers

0.14

Activations Density 0.012%

references to specific brands or figures related to historic and cultural significance

No Comments

No Known Activations