INDEX

Explanations

terms related to hearing and vision

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

usercontent

-0.16

elage

-0.16

ture

-0.15

adr

-0.15

bracht

-0.14

 Arth

-0.14

chem

-0.14

 Fury

-0.14

yre

-0.14

utzer

-0.14

POSITIVE LOGITS

 loss

0.35

 Loss

0.28

aid

0.28

-loss

0.27

loss

0.25

 LOSS

0.25

aid

0.24

Aid

0.24

Loss

0.24

 impairment

0.24

Activations Density 0.009%

terms related to hearing and vision

No Comments

No Known Activations