INDEX

Explanations

structured data elements, particularly related to identification and logging within code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

ÑĸÐ½ÑĮ

-0.16

heet

-0.15

asso

-0.15

hey

-0.14

edula

-0.14

Ow

-0.14

RLF

-0.13

Î¸Î¹

-0.13

MUX

-0.13

orget

-0.13

POSITIVE LOGITS

ardown

0.17

èĬ¬

0.16

liers

0.15

UIS

0.15

ipt

0.15

 tÄĽ

0.14

Ã¼cken

0.14

 åĩ

0.14

 Haut

0.14

ãĤ¿ãĥ¼

0.14

Activations Density 0.028%

structured data elements, particularly related to identification and logging within code

No Comments

No Known Activations