INDEX

Explanations

references to media reports and analyses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

RAW

-0.15

itarian

-0.15

ngen

-0.15

bia

-0.15

 meta

-0.15

meta

-0.14

-meta

-0.14

sez

-0.14

positor

-0.14

urovision

-0.14

POSITIVE LOGITS

 Barr

0.27

WS

0.24

Dow

0.24

.ws

0.24

 Wall

0.23

loomberg

0.22

 Financial

0.22

 CNBC

0.22

 bloom

0.21

WS

0.21

Activations Density 0.138%

references to media reports and analyses

No Comments

No Known Activations