INDEX

Explanations

references to motorsport events and participants

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 sadd

-0.14

lette

-0.14

 Boeh

-0.14

 Leban

-0.14

 Bicycle

-0.14

 saddle

-0.13

 éĸ

-0.13

WithString

-0.13

Ãºa

-0.13

POSITIVE LOGITS

 Racing

0.24

lesi

0.16

 Magn

0.16

 racing

0.16

-corner

0.15

 Glock

0.15

 Hulk

0.15

ikk

0.15

 Alonso

0.15

raq

0.15

Activations Density 0.019%

references to motorsport events and participants

No Comments

No Known Activations