INDEX

Explanations

dates and associated numeric values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

 crib

-0.16

rael

-0.15

unnable

-0.15

alone

-0.14

remium

-0.14

isser

-0.14

cri

-0.14

 pest

-0.14

Ð¸ÑĤÑĭ

-0.14

POSITIVE LOGITS

'gc

0.17

íĤ¹

0.15

ÙĦÙĤ

0.15

.look

0.14

 Roof

0.14

ovah

0.13

Ã¦nd

0.13

 masturb

0.13

ãĤ¤ãĥ«

0.13

 EventType

0.13

Activations Density 0.004%

dates and associated numeric values

No Comments

No Known Activations