INDEX

Explanations

terms related to interpretation and its various forms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

ey

-0.20

Ð½

-0.18

readcr

-0.16

¼

-0.16

uled

-0.15

alf

-0.15

strom

-0.15

itary

-0.15

ergy

-0.15

borg

-0.14

POSITIVE LOGITS

ative

0.23

atively

0.21

ationship

0.19

ive

0.19

ters

0.19

ability

0.17

ations

0.17

-language

0.16

ively

0.15

_singleton

0.15

Activations Density 0.020%

terms related to interpretation and its various forms

No Comments

No Known Activations