INDEX

Explanations

repeated actions or occurrences over time

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

ÑģÑİ

-0.15

zÄħ

-0.15

trand

-0.14

imb

-0.14

ispers

-0.14

inge

-0.14

loor

-0.14

eldon

-0.14

apesh

-0.14

POSITIVE LOGITS

 until

0.20

.until

0.18

until

0.17

 step

0.17

 incremental

0.15

 Until

0.15

_until

0.15

fÃ©

0.15

_STEP

0.15

step

0.14

Activations Density 0.094%

repeated actions or occurrences over time

No Comments

No Known Activations