INDEX

Explanations

connectives and conjunctions that link ideas or clauses

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

.scalablytyped

-0.18

äººãģ¯

-0.16

undler

-0.14

 Derived

-0.14

iid

-0.14

groundColor

-0.13

-UA

-0.13

ilmiÅŁtir

-0.13

udp

-0.13

may

-0.13

POSITIVE LOGITS

 whose

0.19

reso

0.17

 able

0.16

 unable

0.15

 nurs

0.15

indo

0.15

 with

0.14

whose

0.14

anka

0.14

 likely

0.14

Activations Density 0.170%

connectives and conjunctions that link ideas or clauses

No Comments

No Known Activations