INDEX

Explanations

occurrences of the word "last" in various contexts, particularly in reference to time

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

/day

-0.20

Day

-0.18

-Day

-0.17

 daytime

-0.16

DAY

-0.15

noon

-0.15

abus

-0.15

iet

-0.15

oppel

-0.15

ano

-0.15

POSITIVE LOGITS

 year

0.28

 week

0.26

-minute

0.21

 month

0.21

ingly

0.20

year

0.20

minute

0.20

erif

0.17

 years

0.17

 night

0.16

Activations Density 0.019%

occurrences of the word "last" in various contexts, particularly in reference to time

No Comments

No Known Activations