INDEX

Explanations

negative or critical sentiment expressed in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

anja

-0.16

Mam

-0.16

reibung

-0.15

mam

-0.15

oka

-0.15

reib

-0.14

é¼ł

-0.14

ig

-0.14

CHA

-0.14

ering

-0.14

POSITIVE LOGITS

odyn

0.17

ASA

0.16

 é«

0.16

EDIA

0.15

Proto

0.15

ondo

0.15

edia

0.14

nez

0.14

Ð»Ð¸Ð½

0.14

zik

0.13

Activations Density 0.003%

negative or critical sentiment expressed in the text

No Comments

No Known Activations