INDEX

Explanations

references to potential hazards and risks in various contexts

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

 ÑĪÑĤÑĥ

-0.14

_HI

-0.13

ppy

-0.12

 thus

-0.12

 ...,

-0.12

 “âĢ¦

-0.12

pii

-0.12

enie

-0.12

POSITIVE LOGITS

igon

0.15

appen

0.14

Kos

0.13

 ÐŁÐ¾Ðº

0.13

sen

0.13

 sonst

0.13

rÃ¡n

0.12

yk

0.12

erse

0.12

ettel

0.12

Activations Density 0.613%

references to potential hazards and risks in various contexts

No Comments

No Known Activations