INDEX

Explanations

references to technical programming terms and elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

fri

-0.15

èį

-0.15

RLF

-0.15

abis

-0.15

TN

-0.14

 concrete

-0.14

æĿ¡

-0.14

_frm

-0.14

erule

-0.14

elles

-0.14

POSITIVE LOGITS

Purple

0.17

 Purple

0.16

pios

0.15

 Kendrick

0.15

.dep

0.15

/assets

0.15

silver

0.14

Äĳ

0.14

 autos

0.14

ace

0.14

Activations Density 0.004%

references to technical programming terms and elements

No Comments

No Known Activations