INDEX

Explanations

references to technical components or elements in a computing context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ening

-0.19

ernels

-0.17

silver

-0.17

ery

-0.16

teenth

-0.16

ingo

-0.15

ë¨¸ëĭĪ

-0.15

/stats

-0.15

sb

-0.15

ï¸ı

-0.15

POSITIVE LOGITS

ized

0.20

ised

0.18

lle

0.16

IZED

0.16

èħ

0.16

al

0.16

lv

0.15

mates

0.15

ifold

0.15

laus

0.15

Activations Density 0.019%

references to technical components or elements in a computing context

No Comments

No Known Activations