INDEX

Explanations

key phrases related to political support and endorsements

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

/the

-0.17

[]

-0.17

what

-0.15

rible

-0.15

innen

-0.14

/The

-0.14

ajar

-0.13

 éĻ

-0.13

ijd

-0.13

ÅĻÃŃd

-0.13

POSITIVE LOGITS

 same

0.31

own

0.31

 latest

0.29

 entire

0.27

latest

0.23

same

0.22

 ability

0.22

 biggest

0.21

 original

0.20

 newest

0.20

Activations Density 1.240%

key phrases related to political support and endorsements

No Comments

No Known Activations