INDEX

Explanations

symbols, punctuation, and formatting elements 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 @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

udder

-0.17

dan

-0.14

jal

-0.14

ansom

-0.13

htable

-0.13

 devote

-0.13

udd

-0.13

imb

-0.13

alsy

-0.13

onavir

-0.13

POSITIVE LOGITS

 Finger

0.18

OTOS

0.16

 finger

0.15

_Target

0.14

 Swinger

0.14

INET

0.14

finger

0.14

utow

0.14

à¸«à¸§

0.14

Activations Density 0.006%

symbols, punctuation, and formatting elements in the text

No Comments

No Known Activations