INDEX

Explanations

marketing-related phrases and terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

Nil

-0.15

iri

-0.15

ancias

-0.14

 valleys

-0.14

akte

-0.14

chk

-0.13

ÏĨÎ¹

-0.13

.pivot

-0.13

Stamp

-0.13

ergus

-0.13

POSITIVE LOGITS

 signs

0.63

 sign

0.56

 Signs

0.54

 signage

0.47

 Sign

0.43

sign

0.43

Sign

0.41

_sign

0.37

-sign

0.36

.sign

0.35

Activations Density 0.310%

marketing-related phrases and terminology

No Comments

No Known Activations