INDEX

Explanations

references to military and organizational structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

SKTOP

-0.15

inois

-0.15

Them

-0.14

ibri

-0.14

á»ģ

-0.14

icom

-0.14

agues

-0.14

Ð»Ð°Ð³

-0.13

icator

-0.13

POSITIVE LOGITS

Tiny

0.16

 tiny

0.16

 Tiny

0.15

emann

0.15

.boot

0.15

urat

0.14

 Gros

0.14

ÃĹ↵↵

0.14

 permanently

0.14

anos

0.13

Activations Density 0.127%

references to military and organizational structures

No Comments

No Known Activations