INDEX

Explanations

mentions of speaking or expressing opinions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

aml

-0.16

ated

-0.15

om

-0.15

uv

-0.15

ment

-0.15

roit

-0.15

uw

-0.14

reet

-0.14

 fully

-0.14

Äħ

-0.14

POSITIVE LOGITS

 volumes

0.27

 fluent

0.18

ertest

0.17

Volumes

0.15

spe

0.15

 volume

0.15

 engagements

0.15

Ð»Ð°Ð½Ð´

0.15

olumes

0.14

 louder

0.14

Activations Density 0.028%

mentions of speaking or expressing opinions

No Comments

No Known Activations