INDEX

Explanations

code-related elements, particularly focusing on commands and symbols in programming contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

Ã½Å¡

-0.15

aginator

-0.15

beros

-0.15

blast

-0.14

IRA

-0.14

FN

-0.14

_RAD

-0.14

YRO

-0.13

oyo

-0.13

åº

-0.13

POSITIVE LOGITS

ugi

0.17

ific

0.14

omat

0.14

ereg

0.14

uchs

0.13

iek

0.13

ÑĪÐ¸Ð²

0.13

 legacy

0.13

Legacy

0.13

anca

0.13

Activations Density 0.023%

code-related elements, particularly focusing on commands and symbols in programming contexts

No Comments

No Known Activations