INDEX

Explanations

technical terms related to user interface interactions

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

ifix

-0.17

ospace

-0.15

HeaderCode

-0.15

à¸´à¹ī

-0.14

.datatables

-0.14

aliz

-0.14

-fontawesome

-0.14

.opensource

-0.14

ideographic

-0.13

stadt

-0.13

POSITIVE LOGITS

 selection

0.78

 Selection

0.73

 selected

0.67

 selections

0.67

Selection

0.65

selection

0.64

 sele

0.62

-selection

0.62

sel

0.60

 Selected

0.60

Activations Density 0.176%

technical terms related to user interface interactions

No Comments

No Known Activations