INDEX

Explanations

terms related to architecture and architectural professions

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

ascar

-0.17

foy

-0.15

adoras

-0.15

aco

-0.15

eno

-0.14

ØªØ§

-0.14

adora

-0.14

mond

-0.14

ibase

-0.14

ecom

-0.14

POSITIVE LOGITS

kt

0.39

kte

0.23

ckt

0.22

ktion

0.21

KT

0.21

ktop

0.21

ÐºÑĤ

0.21

0.20

ÐºÑĤÑĥ

0.20

uktur

0.19

Activations Density 0.010%

terms related to architecture and architectural professions

No Comments

No Known Activations