INDEX

Explanations

phrases indicating probabilities or likelihoods of events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

 allegedly

-0.23

 supposedly

-0.22

 possibly

-0.18

 presumably

-0.17

 apparently

-0.17

 Apparently

-0.16

 Possibly

-0.16

 arguably

-0.16

 reportedly

-0.16

.au

-0.16

POSITIVE LOGITS

hood

0.41

 hood

0.28

 gonna

0.27

going

0.26

 going

0.26

 candidates

0.26

be

0.24

 scenario

0.22

ly

0.21

 scenarios

0.21

Activations Density 0.044%

phrases indicating probabilities or likelihoods of events

No Comments

No Known Activations