INDEX

Explanations

terms related to network infrastructure and telecommunications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

/operators

-0.16

vens

-0.14

 Hart

-0.14

-0.13

rocessing

-0.13

 camper

-0.13

Cum

-0.13

 ÑģÑĭ

-0.13

éĥİ

-0.13

ennie

-0.13

POSITIVE LOGITS

Opt

0.24

Opt

0.20

Fib

0.20

 optic

0.20

 Coalition

0.20

opt

0.19

 fibre

0.19

 Retail

0.19

Fo

0.19

OPT

0.19

Activations Density 0.007%

terms related to network infrastructure and telecommunications

No Comments

No Known Activations