INDEX

Explanations

code-related keywords and constructs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

guild

-0.17

allee

-0.15

riday

-0.15

æ®Ĭ

-0.15

 tipping

-0.14

ROTO

-0.14

 Guild

-0.14

ErrorException

-0.14

McG

-0.13

Ð¿ÑĢÐ¸ÐºÐ»Ð°Ð´

-0.13

POSITIVE LOGITS

_periods

0.19

tracked

0.18

 Programme

0.17

ãĥ«ãĥĪ

0.17

 organisation

0.17

organisation

0.17

 Program

0.16

 program

0.16

 programme

0.16

 Organisation

0.16

Activations Density 0.002%

code-related keywords and constructs

No Comments

No Known Activations