INDEX

Explanations

instances of question-and-answer dialogues or formats

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

ACEMENT

-0.16

ONSE

-0.15

-0.14

ORAGE

-0.14

coln

-0.14

ector

-0.14

ACING

-0.13

AGON

-0.13

fol

-0.13

LETE

-0.13

POSITIVE LOGITS

uito

0.31

UINT

0.27

&A

0.26

uid

0.26

ues

0.24

ued

0.24

UIT

0.24

iu

0.23

uids

0.23

antas

0.22

Activations Density 0.019%

instances of question-and-answer dialogues or formats

No Comments

No Known Activations