INDEX

Explanations

mentions of inline elements and templates in code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

erland

-0.17

idy

-0.15

icy

-0.14

 Levine

-0.14

 Retro

-0.14

rez

-0.13

rent

-0.13

nt

-0.13

jon

-0.13

 contr

-0.13

POSITIVE LOGITS

ift

0.17

orget

0.15

opts

0.15

heim

0.15

EAR

0.14

ellow

0.14

bios

0.14

 sÃ¡ng

0.14

rana

0.14

burgh

0.14

Activations Density 0.004%

mentions of inline elements and templates in code

No Comments

No Known Activations