INDEX

Explanations

concepts related to social networks and human interactions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

.subscribe

-0.14

ÏĨÏħ

-0.13

pek

-0.13

 rozh

-0.13

.infinity

-0.13

ordinate

-0.13

 disparate

-0.13

nav

-0.13

eyin

-0.13

 Jenner

-0.12

POSITIVE LOGITS

 network

0.24

 networks

0.20

network

0.20

 patterns

0.20

 social

0.19

 Sadd

0.17

 activity

0.17

 Patterns

0.17

 Network

0.16

 Social

0.16

Activations Density 0.146%

concepts related to social networks and human interactions

No Comments

No Known Activations