INDEX

Explanations

terms related to conflict, prisoners, and government surveillance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

IRT

-0.16

LEC

-0.16

NET

-0.15

amu

-0.15

anium

-0.14

æĪ

-0.14

 Gent

-0.14

ledo

-0.14

 spot

-0.14

RIA

-0.14

POSITIVE LOGITS

 Ø¯Ø§Ùħ

0.16

enberg

0.15

opsis

0.15

abee

0.15

linkplain

0.14

phem

0.14

nton

0.14

antas

0.14

defer

0.14

Ð¶Ð¸

0.14

Activations Density 0.013%

terms related to conflict, prisoners, and government surveillance

No Comments

No Known Activations