INDEX

Explanations

phrases indicating relationships and correspondences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

utr

-0.17

rah

-0.15

-0.14

âĪ

-0.14

lice

-0.14

sv

-0.14

 Tato

-0.14

ll

-0.13

sp

-0.13

POSITIVE LOGITS

 nuru

0.17

activex

0.16

abol

0.16

enha

0.16

xbd

0.15

TriState

0.15

ãĥ³ãĤ¬

0.15

ingly

0.15

DMIN

0.14

-sex

0.14

Activations Density 0.026%

phrases indicating relationships and correspondences

No Comments

No Known Activations