INDEX

Explanations

significant mentions of characters, events, or elements related to a narrative storyline

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Carnegie

-0.16

ëĮĢë¥¼

-0.15

CAS

-0.15

 Bull

-0.15

olo

-0.15

enheim

-0.15

ibil

-0.15

inge

-0.14

 Bowl

-0.14

POSITIVE LOGITS

 Grimm

0.22

 Beacon

0.20

 Jaune

0.19

 Monroe

0.19

 Weiss

0.19

itsu

0.18

 Blake

0.17

Atlas

0.17

 Atlas

0.17

 Yang

0.17

Activations Density 0.001%

significant mentions of characters, events, or elements related to a narrative storyline

No Comments

No Known Activations