INDEX

Explanations

technical terms and concepts related to physics and engineering

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

 Tune

-0.14

ospace

-0.14

ivi

-0.14

caster

-0.14

 Weed

-0.14

 beat

-0.14

 powder

-0.13

pread

-0.13

ë²Į

-0.13

civ

-0.13

POSITIVE LOGITS

 boundary

0.43

 Boundary

0.37

boundary

0.37

Boundary

0.35

_boundary

0.32

BC

0.28

 Robin

0.28

 boundaries

0.27

Dir

0.27

BC

0.25

Activations Density 0.122%

technical terms and concepts related to physics and engineering

No Comments

No Known Activations