INDEX

Explanations

concepts related to exploration and depth in various contexts, such as space, color, and personal relationships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

hurst

-0.16

ollen

-0.15

 Subway

-0.15

 Sunder

-0.14

piler

-0.14

 Sustainability

-0.14

pone

-0.14

yleft

-0.13

ÐµÐ½ÑĮ

-0.13

à¸²à¸į

-0.13

POSITIVE LOGITS

 space

0.96

space

0.80

-space

0.77

 Space

0.76

_space

0.72

Space

0.71

ç©ºéĹ´

0.68

 espacio

0.65

 SPACE

0.65

 spaces

0.63

Activations Density 0.249%

concepts related to exploration and depth in various contexts, such as space, color, and personal relationships

No Comments

No Known Activations