INDEX

Explanations

phrases related to preparation and planning

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

nd

-0.18

nds

-0.18

leans

-0.17

ediator

-0.16

fy

-0.15

ity

-0.15

hed

-0.14

nde

-0.14

deaux

-0.14

faction

-0.14

POSITIVE LOGITS

quarters

0.20

ward

0.17

ernet

0.17

0.16

quarter

0.15

butt

0.15

wards

0.15

Activations Density 0.013%

phrases related to preparation and planning

No Comments

No Known Activations