INDEX

Explanations

references to time periods and frequencies related to work and activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

quo

-0.16

erras

-0.16

uais

-0.15

Years

-0.14

etri

-0.14

zah

-0.13

years

-0.13

IRD

-0.13

atis

-0.13

urch

-0.13

POSITIVE LOGITS

 week

0.82

week

0.68

 Week

0.66

 weeks

0.65

-week

0.64

 WEEK

0.61

Week

0.59

_week

0.59

.week

0.57

weeks

0.55

Activations Density 0.112%

references to time periods and frequencies related to work and activities

No Comments

No Known Activations