INDEX

Explanations

specific symbols or characters indicating encoding issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

oho

-0.16

_plural

-0.15

Ã¡nd

-0.15

_TA

-0.14

 Alliance

-0.14

achat

-0.14

áº¡n

-0.14

ÑĪÐ°

-0.14

 permalink

-0.14

TA

-0.14

POSITIVE LOGITS

 Face

0.29

 FACE

0.29

 face

0.29

Face

0.26

 faces

0.26

-face

0.26

_face

0.26

FACE

0.25

 CLASS

0.25

face

0.24

Activations Density 0.003%

specific symbols or characters indicating encoding issues

No Comments

No Known Activations