INDEX

Explanations

technical terms and references related to programming and libraries

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

 omas

-0.16

éª

-0.15

emark

-0.14

Assembler

-0.14

 Hour

-0.14

rika

-0.14

asmus

-0.14

Hour

-0.14

ULER

-0.14

ela

-0.14

POSITIVE LOGITS

 stable

0.15

obo

0.15

 zost

0.15

lÃ½

0.14

vie

0.14

 complex

0.14

obar

0.13

SI

0.13

fake

0.13

oux

0.13

Activations Density 0.075%

technical terms and references related to programming and libraries

No Comments

No Known Activations