INDEX

Explanations

elements related to specific places and their details

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

ingham

-0.17

Å¾el

-0.16

eda

-0.15

mod

-0.15

ITCH

-0.14

elor

-0.14

zÄħ

-0.14

-sem

-0.14

Å¾il

-0.13

POSITIVE LOGITS

ÏĥÎ¹Î¿

0.15

adoo

0.15

ëģ

0.15

gaard

0.14

anco

0.14

 Guild

0.14

oir

0.14

shi

0.14

çĶ

0.14

Ø¶ÙĦ

0.13

Activations Density 0.023%

elements related to specific places and their details

No Comments

No Known Activations