INDEX

Explanations

terms related to software development or programming environments

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

vak

-0.15

abeth

-0.14

nger

-0.14

rÃ½

-0.13

alfa

-0.13

onu

-0.13

pel

-0.13

_kw

-0.13

 Blake

-0.13

oplast

-0.13

POSITIVE LOGITS

 Progress

0.24

.progress

0.23

 progress

0.22

Progress

0.20

progress

0.20

-progress

0.17

(progress

0.16

cco

0.16

 progresses

0.15

.project

0.15

Activations Density 0.005%

terms related to software development or programming environments

No Comments

No Known Activations