INDEX

Explanations

concepts related to principles and comparisons among different entities

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

ifar

-0.16

elyn

-0.15

endar

-0.14

icari

-0.14

uilt

-0.14

ahi

-0.13

asse

-0.13

 Tibetan

-0.13

aurant

-0.13

kir

-0.13

POSITIVE LOGITS

 apply

0.77

 applies

0.73

apply

0.71

 applic

0.68

 Apply

0.67

 applied

0.67

Apply

0.67

.apply

0.66

 applicable

0.63

 applying

0.63

Activations Density 0.274%

concepts related to principles and comparisons among different entities

No Comments

No Known Activations