INDEX

Explanations

code-related keywords and programming constructs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

 çĬ

-0.16

IB

-0.15

 multiplic

-0.14

.enums

-0.14

 Ethnic

-0.14

 intensity

-0.14

Intensity

-0.14

opic

-0.14

rick

-0.13

 towers

-0.13

POSITIVE LOGITS

ilot

0.17

 dialect

0.16

SSIP

0.16

uela

0.15

unan

0.15

NAN

0.15

ãĥīãĥ«

0.14

zahl

0.14

 Banc

0.14

preter

0.14

Activations Density 0.169%

code-related keywords and programming constructs

No Comments

No Known Activations