INDEX

Explanations

references to images and their credits in the text

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

 ActionTypes

-0.16

æħİ

-0.16

SSF

-0.15

 tune

-0.15

ubat

-0.15

Ø¹ÙĨÙĪØ§ÙĨ

-0.14

unta

-0.14

erson

-0.14

izens

-0.14

gil

-0.14

POSITIVE LOGITS

PA

0.35

PA

0.32

pa

0.22

_PA

0.21

Mirror

0.19

pa

0.19

 Mirror

0.18

 PRESS

0.18

.pa

0.18

Pa

0.18

Activations Density 0.015%

references to images and their credits in the text

No Comments

No Known Activations