INDEX

Explanations

references to technology and programming concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

ÐºÑĸÐ½

-0.15

 Transparent

-0.14

eldorf

-0.14

junction

-0.14

uddy

-0.14

uten

-0.14

acco

-0.14

arro

-0.14

phy

-0.14

 Anders

-0.14

POSITIVE LOGITS

 metric

0.23

 Metric

0.23

 metrics

0.23

 traces

0.22

 otel

0.21

Metric

0.21

 exporters

0.21

 Prometheus

0.21

 exporter

0.20

 instrument

0.20

Activations Density 0.012%

references to technology and programming concepts

No Comments

No Known Activations