INDEX

Explanations

comments and documentation sections in code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

led

-0.14

atura

-0.14

owned

-0.14

ledi

-0.14

@testable

-0.13

caf

-0.13

ãģĭãĤīãģ¯

-0.13

inem

-0.13

èģ

-0.13

inus

-0.13

POSITIVE LOGITS

oki

0.20

ë¹Ī

0.16

LogLevel

0.16

emey

0.15

Ã¤nde

0.15

 Trojan

0.14

fir

0.14

ìĩ

0.14

 Ð´Ð¾ÑĤ

0.14

ency

0.13

Activations Density 0.036%

comments and documentation sections in code

No Comments

No Known Activations