INDEX

Explanations

references to personal perspectives and energies associated with various subjects

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

kar

-0.18

 addCriterion

-0.17

ë¦¬ìĸ´

-0.15

èĦĤ

-0.15

 Kapoor

-0.15

.synthetic

-0.15

ncpy

-0.14

idar

-0.14

addir

-0.14

bundles

-0.14

POSITIVE LOGITS

ze

0.16

awl

0.16

yc

0.16

VED

0.15

xa

0.15

iek

0.14

cs

0.14

ile

0.14

eci

0.14

omial

0.13

Activations Density 0.551%

references to personal perspectives and energies associated with various subjects

No Comments

No Known Activations