INDEX

Explanations

references to hate crimes and violent offenses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

imli

-0.16

ghan

-0.16

 Pitch

-0.16

agrant

-0.15

MMdd

-0.15

.pitch

-0.15

Escort

-0.15

.sg

-0.14

Ø±ÙĪØ¬

-0.14

ugs

-0.14

POSITIVE LOGITS

ä»²

0.16

çĶ¨åĵģ

0.16

.ReadString

0.14

 contr

0.14

åĩĮ

0.14

0.13

 Messenger

0.13

Desk

0.13

iband

0.13

USH

0.13

Activations Density 0.005%

references to hate crimes and violent offenses

No Comments

No Known Activations