INDEX

Explanations

words and terms related to technology or engineering components

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

tk

-0.16

itez

-0.15

et

-0.15

exels

-0.15

itos

-0.15

itus

-0.15

-0.14

TK

-0.14

LC

-0.14

oir

-0.14

POSITIVE LOGITS

ilded

0.20

hey

0.19

ues

0.17

ateway

0.17

apers

0.17

hear

0.17

inning

0.16

he

0.16

orda

0.16

ediÄŁi

0.16

Activations Density 0.045%

words and terms related to technology or engineering components

No Comments

No Known Activations