INDEX

Explanations

words related to preferences and choices

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

ercul

-0.17

inea

-0.16

angelo

-0.16

suming

-0.15

quee

-0.15

aba

-0.14

mand

-0.14

rom

-0.14

igh

-0.14

rapy

-0.14

POSITIVE LOGITS

entially

0.35

ential

0.30

ably

0.20

atory

0.17

/pre

0.17

ett

0.16

NCY

0.16

.Preference

0.16

ongyang

0.15

ited

0.15

Activations Density 0.033%

words related to preferences and choices

No Comments

No Known Activations