INDEX

Explanations

technical coding terms and references related to software development

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

erland

-0.16

.tt

-0.14

achi

-0.14

acman

-0.13

iani

-0.13

 Wrath

-0.13

asley

-0.13

Î»Ïī

-0.12

 Corner

-0.12

orpor

-0.12

POSITIVE LOGITS

urer

0.13

ãĥĥ

0.13

emi

0.13

Quad

0.13

 double

0.13

 Scout

0.12

altura

0.12

 DataColumn

0.12

_overlap

0.12

ILTER

0.12

Activations Density 0.143%

technical coding terms and references related to software development

No Comments

No Known Activations