INDEX

Explanations

references to quantities and counts

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

axe

-0.16

.opendaylight

-0.15

 Weeks

-0.15

anton

-0.14

sons

-0.14

ary

-0.13

 Months

-0.13

ãĤĪãģĨãģª

-0.13

oil

-0.13

shint

-0.13

POSITIVE LOGITS

 times

0.71

 Times

0.55

times

0.53

 TIMES

0.52

Times

0.49

_times

0.45

-times

0.43

 veces

0.41

 fois

0.41

 vezes

0.39

Activations Density 0.049%

references to quantities and counts

No Comments

No Known Activations