INDEX

Explanations

occurrences of the word "Title" followed by numbers, indicating different sections or topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 19-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L19R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.19.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

oday

-0.16

ery

-0.16

aby

-0.16

ermann

-0.15

epad

-0.15

elyn

-0.15

lek

-0.14

iness

-0.14

our

-0.14

POSITIVE LOGITS

tit

0.20

=title

0.19

>Title

0.17

holders

0.16

 deed

0.15

aal

0.15

phoon

0.15

.authenticate

0.15

PREFIX

0.15

pond

0.15

Activations Density 0.029%

occurrences of the word "Title" followed by numbers, indicating different sections or topics

No Comments

No Known Activations