INDEX

Explanations

references to mathematical methods or models associated with research studies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

Ã¹a

-0.19

aven

-0.18

edi

-0.17

IME

-0.17

asons

-0.17

aker

-0.16

ime

-0.16

ensch

-0.16

ensa

-0.16

ama

-0.16

POSITIVE LOGITS

ched

0.24

cc

0.19

ee

0.18

roz

0.17

eller

0.17

abit

0.16

cd

0.16

ard

0.16

icle

0.16

lod

0.16

Activations Density 0.040%

references to mathematical methods or models associated with research studies

No Comments

No Known Activations