INDEX

Explanations

punctuation marks, particularly periods and commas, indicating sentence and clause boundaries 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 @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

 Bloss

-0.17

bis

-0.16

ermen

-0.15

.communication

-0.15

orex

-0.14

ÏĢÎ»

-0.14

IEL

-0.14

ependency

-0.14

çĢ

-0.13

ullen

-0.13

POSITIVE LOGITS

aus

0.15

lightbox

0.14

ha

0.14

prompt

0.14

 Court

0.14

pod

0.14

 ,č↵

0.13

halt

0.13

 Finger

0.13

ëıĻ

0.13

Activations Density 0.007%

punctuation marks, particularly periods and commas, indicating sentence and clause boundaries in the text

No Comments

No Known Activations