INDEX

Explanations

repeated characters or patterns in text

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

ete

-0.16

avy

-0.16

owa

-0.15

enko

-0.15

opa

-0.15

ering

-0.15

aging

-0.15

oci

-0.15

awa

-0.15

ings

-0.14

POSITIVE LOGITS

ÑĥÐ´Ð¸

0.25

ÑĢÐ³

0.20

Ð½Ð°

0.19

Ð»ÑĮÑĤ

0.17

ÑĥÐº

0.17

trak

0.17

ÑĢÐ¸ÑģÑĤ

0.17

Ð¼Ð¾ÑĢ

0.16

Ð½Ð¸Ð¼

0.16

ÑĢÐ°Ð±

0.16

Activations Density 0.009%

repeated characters or patterns in text

No Comments

No Known Activations