INDEX

Explanations

phrases or symbols related to code execution or function calls

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

rum

-0.15

IGHL

-0.15

wick

-0.15

dy

-0.14

 ë§ŀ

-0.14

beit

-0.14

(Syntax

-0.13

ARRIER

-0.13

inning

-0.13

ually

-0.13

POSITIVE LOGITS

alet

0.20

âĨĴâĨĴ

0.17

erken

0.15

ãģĬãĤĬ

0.14

 dÃµi

0.14

ÃŃrk

0.14

Fri

0.14

appy

0.13

presso

0.13

Activations Density 0.036%

phrases or symbols related to code execution or function calls

No Comments

No Known Activations