INDEX

Explanations

connections and interactions among characters or individuals in a narrative

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

onda

-0.15

èĬ¯

-0.15

awi

-0.15

 nÃło

-0.14

izon

-0.14

Miz

-0.14

Writes

-0.14

 himself

-0.13

 Solomon

-0.13

ernetes

-0.13

POSITIVE LOGITS

pector

0.16

gether

0.16

iado

0.14

 kork

0.14

darwin

0.14

antan

0.14

amel

0.14

 birlikte

0.14

 ninh

0.13

ä¸¤äºº

0.13

Activations Density 0.400%

connections and interactions among characters or individuals in a narrative

No Comments

No Known Activations