INDEX

Explanations

verbs and forms of action related to doing

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

 simply

-0.19

 simplement

-0.17

ãĥ¼ãĥĨãĤ£

-0.15

 straightforward

-0.15

 plain

-0.15

ibern

-0.15

 simples

-0.15

alar

-0.15

itor

-0.14

 mere

-0.14

POSITIVE LOGITS

ju

0.44

Ju

0.34

ju

0.34

Ju

0.31

ust

0.30

 jest

0.29

0.28

ust

0.28

js

0.27

(j

0.25

Activations Density 0.076%

verbs and forms of action related to doing

No Comments

No Known Activations