INDEX

Explanations

references to political leaders and officials

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

agi

-0.15

.Guna

-0.15

âĸį

-0.15

emmel

-0.15

itos

-0.14

ALLE

-0.14

ÏģÎŃ

-0.14

untas

-0.14

ofile

-0.14

ÑĥÐº

-0.14

POSITIVE LOGITS

 spokesman

0.19

 spokeswoman

0.19

 spokesperson

0.17

 said

0.15

on

0.15

sp

0.15

Wyn

0.15

on

0.15

Ľå»º

0.15

 Pink

0.14

Activations Density 0.118%

references to political leaders and officials

No Comments

No Known Activations