INDEX

Explanations

keywords related to programming functions and structure

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

th

-0.19

hil

-0.16

alt

-0.16

 Dess

-0.15

ar

-0.15

edit

-0.15

ct

-0.14

nock

-0.14

mens

-0.14

strings

-0.14

POSITIVE LOGITS

UNUSED

0.17

ToLocal

0.16

aland

0.16

iÄįka

0.16

umberland

0.15

raki

0.15

iska

0.15

xfff

0.15

umba

0.14

zeÅĦ

0.14

Activations Density 0.004%

keywords related to programming functions and structure

No Comments

No Known Activations