INDEX

Explanations

symbols or characters used in URLs or code formatting

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

-0.17

 gate

-0.14

Kar

-0.14

aÃ±

-0.14

wax

-0.14

-0.13

POSITIVE LOGITS

 pylint

0.18

noinspection

0.17

_:*

0.16

Î¹Î±Î½

0.15

isphere

0.15

ephir

0.15

okit

0.15

.changed

0.14

íķĻê¸°

0.14

bsolute

0.14

Activations Density 0.013%

symbols or characters used in URLs or code formatting

No Comments

No Known Activations