INDEX

Explanations

words related to legal or formal statements and their nuances

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

dat

-0.18

Dat

-0.17

th

-0.15

 than

-0.14

 hats

-0.14

ÐµÐ¸

-0.14

ering

-0.14

 Than

-0.13

_fds

-0.13

 Query

-0.13

POSITIVE LOGITS

que

0.83

Que

0.55

Que

0.55

_que

0.53

QUE

0.51

qu

0.50

que

0.50

 queue

0.40

.queue

0.38

 queues

0.37

Activations Density 0.054%

words related to legal or formal statements and their nuances

No Comments

No Known Activations