INDEX

Explanations

technical terms related to programming and compilation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

.scalablytyped

-0.18

isme

-0.14

clus

-0.14

acy

-0.14

tgl

-0.13

UDP

-0.13

Î·Î½

-0.13

ÚĺÙĩ

-0.13

emo

-0.13

aster

-0.13

POSITIVE LOGITS

 symbol

0.32

sym

0.31

symbol

0.30

 symbols

0.30

 Symbols

0.29

sym

0.29

elf

0.28

.sym

0.28

symbols

0.28

 SYMBOL

0.28

Activations Density 0.046%

technical terms related to programming and compilation

No Comments

No Known Activations