INDEX

Explanations

references to specific racing events and achievements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 3-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L3R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.3.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

zev

-0.17

roje

-0.17

ÛĮÙĨÙĩ

-0.16

jal

-0.15

phe

-0.15

ãĤ¤ãĤº

-0.14

arella

-0.14

oux

-0.14

ednou

-0.14

icles

-0.13

POSITIVE LOGITS

isma

0.16

lint

0.15

 hypo

0.15

andin

0.14

 quot

0.14

bat

0.14

uhan

0.14

etro

0.14

Dag

0.14

 super

0.14

Activations Density 0.473%

references to specific racing events and achievements

No Comments

No Known Activations