INDEX

Explanations

references to positions of political leadership, particularly vice presidents and their mentions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

amu

-0.15

 Ø¨ÙĪØ§Ø¨Ø©

-0.14

_BO

-0.14

å·

-0.14

emacs

-0.14

ĵ°

-0.13

 ç¥ŀ

-0.13

incinn

-0.13

ao

-0.13

assed

-0.13

POSITIVE LOGITS

pond

0.17

ãĤ´ãĥª

0.15

phylum

0.15

gii

0.15

hone

0.15

eum

0.14

ennen

0.14

cog

0.14

olland

0.14

 undermin

0.14

Activations Density 0.009%

references to positions of political leadership, particularly vice presidents and their mentions

No Comments

No Known Activations