INDEX

Explanations

code related to programming constructs and data manipulation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

pir

-0.15

erdale

-0.14

hatt

-0.14

Ł³

-0.14

dag

-0.14

ikat

-0.14

icken

-0.14

dir

-0.14

API

-0.13

ÄħÅ¼

-0.13

POSITIVE LOGITS

 split

0.29

.Split

0.29

.split

0.27

 splits

0.26

 Split

0.26

split

0.25

 splitting

0.25

-split

0.25

Split

0.25

 components

0.24

Activations Density 0.128%

code related to programming constructs and data manipulation

No Comments

No Known Activations