INDEX

Explanations

references to specific institutions or entities, particularly with the word "ant" appearing frequently in various contexts

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

-0.31

ska

-0.23

-0.22

-0.21

à¤°

-0.21

hape

-0.20

rms

-0.20

spr

-0.20

iÅ¡

-0.20

POSITIVE LOGITS

ech

0.32

0.28

ucket

0.27

yh

0.26

ec

0.26

elope

0.26

astic

0.25

eg

0.23

ing

0.22

yne

0.22

Activations Density 0.024%

references to specific institutions or entities, particularly with the word "ant" appearing frequently in various contexts

No Comments

No Known Activations