INDEX

Explanations

phrases that express exactness or specificity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

ye

-0.18

unts

-0.16

rella

-0.15

quist

-0.15

entic

-0.15

ess

-0.15

exion

-0.15

oran

-0.14

 specifically

-0.14

ping

-0.14

POSITIVE LOGITS

itude

0.26

 opposite

0.20

itudes

0.17

ITUDE

0.17

-ÑĤÐ°ÐºÐ¸

0.16

aleb

0.16

elly

0.16

à¹Ħà¸«à¸Ļ

0.16

;y

0.16

ingly

0.15

Activations Density 0.025%

phrases that express exactness or specificity

No Comments

No Known Activations