INDEX

Explanations

references to death, killing, and violence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

PGA

-0.15

 deadly

-0.14

 tear

-0.14

loha

-0.14

$__

-0.14

-0.13

ida

-0.13

ifetime

-0.13

 damage

-0.13

POSITIVE LOGITS

olson

0.16

æİī

0.16

 innocent

0.15

ÏģÎŃ

0.15

 INCIDENTAL

0.14

æº

0.14

ivet

0.14

ÏģÎµ

0.14

defense

0.14

 hung

0.14

Activations Density 0.159%

references to death, killing, and violence

No Comments

No Known Activations