INDEX

Explanations

numerical data and technical specifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

enor

-0.16

privation

-0.16

wright

-0.15

ÙİØ§

-0.15

 ÐºÑĢÐ¸

-0.14

uron

-0.14

rog

-0.14

ureka

-0.14

POSITIVE LOGITS

dorf

0.17

Decorator

0.15

odb

0.14

_decorator

0.14

oll

0.14

ë£¡

0.13

ansk

0.13

 Execution

0.13

nes

0.13

ANS

0.13

Activations Density 0.284%

numerical data and technical specifications

No Comments

No Known Activations