INDEX

Explanations

references to historical events or organizations related to war and memorials

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

ipsis

-0.16

ips

-0.15

 themselves

-0.15

search

-0.15

''

-0.14

 vice

-0.14

etc

-0.14

-d

-0.14

POSITIVE LOGITS

 consists

0.18

ÌĨ

0.17

 contains

0.17

 consist

0.16

ythe

0.15

contains

0.15

Ymd

0.15

is

0.15

ï¼Įå®ĥ

0.15

.intellij

0.15

Activations Density 0.217%

references to historical events or organizations related to war and memorials

No Comments

No Known Activations