INDEX

Explanations

references to racing events and motorsport achievements

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

æĪ²

-0.15

quate

-0.14

 softball

-0.14

 à¤¤à¤²

-0.14

 volleyball

-0.14

imals

-0.13

igua

-0.13

roid

-0.13

 Lehr

-0.13

 à¤Ŀ

-0.13

POSITIVE LOGITS

 race

0.50

 Race

0.46

 races

0.44

 racing

0.44

Race

0.42

race

0.41

 racer

0.39

rac

0.39

 drivers

0.38

 driver

0.38

Activations Density 0.347%

references to racing events and motorsport achievements

No Comments

No Known Activations