INDEX

Explanations

references to scientific methodologies and data analysis techniques

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

htm

-0.16

brtc

-0.14

 tear

-0.14

 tears

-0.13

-valu

-0.13

arih

-0.13

aton

-0.13

 ekip

-0.13

DTV

-0.13

dk

-0.13

POSITIVE LOGITS

 sequences

0.32

 sequence

0.30

 Sequence

0.29

Seq

0.27

NC

0.25

 fasta

0.24

 FAST

0.24

Tax

0.24

 blast

0.24

sequence

0.24

Activations Density 0.048%

references to scientific methodologies and data analysis techniques

No Comments

No Known Activations