INDEX

Explanations

historical references to military events and actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

 NATO

-0.18

Lob

-0.16

Vad

-0.15

Ãª

-0.15

 tuto

-0.14

 Montreal

-0.14

hem

-0.14

iero

-0.14

_errno

-0.14

 Gaut

-0.14

POSITIVE LOGITS

 Pearl

0.26

 Pacific

0.25

 Philippine

0.23

 Guam

0.23

Pacific

0.22

-Pacific

0.21

Pear

0.21

okus

0.20

Luz

0.20

 Filipino

0.20

Activations Density 0.029%

historical references to military events and actions

No Comments

No Known Activations