INDEX

Explanations

names and identifiers related to people and entities

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

byname

-0.17

/tos

-0.17

CallCheck

-0.16

_tA

-0.15

.mdl

-0.15

chio

-0.15

?family

-0.15

ederland

-0.15

.ecore

-0.14

erosis

-0.14

POSITIVE LOGITS

Software

0.24

 Software

0.23

Soft

0.21

-develop

0.20

coder

0.20

 âĻ

0.19

SOFTWARE

0.19

Developer

0.18

software

0.18

labs

0.18

Activations Density 0.156%

names and identifiers related to people and entities

No Comments

No Known Activations