INDEX

Explanations

phrases indicating possibilities or options

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

 åĳ

-0.15

ertia

-0.14

 arrival

-0.14

agna

-0.14

oky

-0.13

 Alleg

-0.13

 miss

-0.13

 surrounded

-0.13

Alternate

-0.13

icket

-0.12

POSITIVE LOGITS

 connections

0.17

pig

0.17

 crossover

0.16

 Connections

0.15

pig

0.15

ause

0.15

 rencont

0.15

agged

0.15

Connections

0.15

 connection

0.15

Activations Density 0.010%

phrases indicating possibilities or options

No Comments

No Known Activations