INDEX

Explanations

words that indicate transfer or movement between states or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

:CGRect

-0.17

-0.15

à¤Ŀ

-0.15

alen

-0.15

od

-0.14

ynth

-0.14

 Twin

-0.14

ħn

-0.13

POSITIVE LOGITS

udeau

0.15

okie

0.15

Î´ÏģÎ¿

0.14

embros

0.14

 cool

0.14

idal

0.14

BitFields

0.14

ulado

0.14

annels

0.14

rych

0.14

Activations Density 0.069%

words that indicate transfer or movement between states or conditions

No Comments

No Known Activations