INDEX

Explanations

phrases related to misinformation and myths

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

yi

-0.14

 retrospect

-0.14

inger

-0.14

ird

-0.14

 Dysfunction

-0.13

áº¿t

-0.13

.dtd

-0.13

cef

-0.13

achts

-0.13

oder

-0.13

POSITIVE LOGITS

 myth

0.61

 myths

0.59

 Myth

0.56

 mythology

0.42

 falsehood

0.40

 false

0.40

 legend

0.38

 legends

0.38

 Legend

0.35

 mythical

0.35

Activations Density 0.453%

phrases related to misinformation and myths

No Comments

No Known Activations