INDEX

Explanations

critical terms and concepts in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

Tol

-0.17

Tee

-0.15

arella

-0.14

 proh

-0.14

 living

-0.14

dig

-0.14

ieux

-0.14

dig

-0.14

kre

-0.14

endi

-0.13

POSITIVE LOGITS

oeff

0.17

Volumes

0.15

dech

0.15

PRINTF

0.14

uibModal

0.14

ude

0.14

Ø§Ø¡

0.14

ackage

0.14

 ÙģØ§Ø±Ø³ÛĮ

0.14

ched

0.14

Activations Density 0.002%

critical terms and concepts in various contexts

No Comments

No Known Activations