INDEX

Explanations

file and directory structure terms related to project management or code organization

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

ulu

-0.16

 blind

-0.15

IMS

-0.15

vanced

-0.14

.vn

-0.14

dep

-0.14

 Blind

-0.13

[('

-0.13

antom

-0.13

POSITIVE LOGITS

ellt

0.15

UNDLE

0.15

ziej

0.15

uncia

0.15

evin

0.15

æį·

0.14

axe

0.14

sep

0.14

arker

0.14

lehem

0.14

Activations Density 0.062%

file and directory structure terms related to project management or code organization

No Comments

No Known Activations