INDEX

Explanations

references to repeated actions or occurrences

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

abar

-0.18

agua

-0.17

aqu

-0.15

arf

-0.14

zia

-0.14

CP

-0.14

egan

-0.14

Ø±ÙĪØ¶

-0.14

cake

-0.13

auth

-0.13

POSITIVE LOGITS

kowski

0.17

Disappear

0.16

@register

0.16

kker

0.15

uo

0.15

 medicine

0.15

 Fluid

0.15

fluid

0.15

 SolidColorBrush

0.14

 Nice

0.14

Activations Density 0.072%

references to repeated actions or occurrences

No Comments

No Known Activations