INDEX

Explanations

technical jargon and terms related to processes and components in scientific and engineering contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

 itself

-0.20

Ñķ

-0.18

serve

-0.16

lets

-0.16

(s

-0.16

’s

-0.16

(es

-0.16

ss

-0.15

cz

-0.15

hn

-0.15

POSITIVE LOGITS

ickness

0.19

ight

0.18

cales

0.18

ATUS

0.17

ided

0.16

oda

0.16

heets

0.16

 themselves

0.15

aber

0.15

ighting

0.15

Activations Density 4.314%

technical jargon and terms related to processes and components in scientific and engineering contexts

No Comments

No Known Activations