INDEX

Explanations

references to ancient topics and civilizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

backs

-0.17

nic

-0.15

çİĩ

-0.15

recision

-0.14

ero

-0.14

sonian

-0.14

enny

-0.13

à¤¬à¤²

-0.13

oke

-0.13

POSITIVE LOGITS

ly

0.19

-old

0.17

usz

0.15

bolt

0.14

liness

0.14

-looking

0.14

afi

0.14

ollo

0.14

.getSeconds

0.13

ofile

0.13

Activations Density 0.010%

references to ancient topics and civilizations

No Comments

No Known Activations