INDEX

Explanations

occurrences of the token "-st" across various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

rw

-0.25

rut

-0.23

rh

-0.20

rane

-0.20

rÃ©

-0.20

rar

-0.20

rin

-0.20

rig

-0.19

rax

-0.19

POSITIVE LOGITS

udio

0.35

rength

0.34

atement

0.33

roke

0.32

reet

0.32

udy

0.31

rike

0.31

arter

0.30

ories

0.30

ripe

0.30

Activations Density 0.015%

occurrences of the token "-st" across various contexts

No Comments

No Known Activations