INDEX

Explanations

references to popular or highly ranked items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

ntity

-0.18

 (*((

-0.17

enaire

-0.15

ensch

-0.14

TickCount

-0.14

chwitz

-0.14

ridge

-0.14

odka

-0.14

ukkit

-0.13

POSITIVE LOGITS

Ø±Ø¯

0.16

äº

0.15

ereal

0.15

gon

0.15

 list

0.15

0.14

_physical

0.14

 nhau

0.14

 listing

0.14

anter

0.14

Activations Density 0.129%

references to popular or highly ranked items

No Comments

No Known Activations