INDEX

Explanations

references to organizations and recognition of individuals or groups

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

ledi

-0.17

ijd

-0.16

caffe

-0.14

áº¿

-0.14

 Redistributions

-0.14

 Blasio

-0.14

 Packing

-0.14

Occurred

-0.14

pector

-0.14

 Bott

-0.14

POSITIVE LOGITS

 rapport

0.25

Rap

0.25

rap

0.25

EP

0.23

rap

0.21

EP

0.21

EMPL

0.20

MEP

0.20

pl

0.20

 codec

0.19

Activations Density 0.015%

references to organizations and recognition of individuals or groups

No Comments

No Known Activations