INDEX

Explanations

mentions of quantity and variety in context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 Many

-0.17

 someone

-0.17

 anyone

-0.16

 Various

-0.15

rowsable

-0.15

 somebody

-0.15

 something

-0.15

processable

-0.15

OTES

-0.14

POSITIVE LOGITS

/all

0.25

 dozen

0.24

 kinds

0.22

 types

0.20

ãĢħ

0.19

 important

0.19

-thirds

0.19

 pieces

0.18

 facets

0.18

teenth

0.17

Activations Density 0.349%

mentions of quantity and variety in context

No Comments

No Known Activations