INDEX

Explanations

references to individuals with extensive experience or notable credentials

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

Hav

-0.17

atee

-0.15

 Milo

-0.14

 Millet

-0.14

StackSize

-0.14

 Bott

-0.13

 Sunset

-0.13

ITIZE

-0.13

.dm

-0.13

 tabi

-0.13

POSITIVE LOGITS

Ĳ

0.16

blr

0.15

orden

0.15

enda

0.15

ixin

0.15

mint

0.14

anden

0.14

ajor

0.14

outh

0.14

Activations Density 0.004%

references to individuals with extensive experience or notable credentials

No Comments

No Known Activations