INDEX

Explanations

occurrences of the word "in."

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

-0.17

vert

-0.15

ãĥ¥

-0.15

ivel

-0.14

iling

-0.14

anto

-0.14

bat

-0.14

aha

-0.14

etz

-0.14

elt

-0.13

POSITIVE LOGITS

tow

0.31

 sight

0.30

 play

0.25

 reach

0.23

 hand

0.21

 Sight

0.21

 plain

0.19

Evidence

0.18

ighted

0.18

 evidence

0.18

Activations Density 0.161%

occurrences of the word "in."

No Comments

No Known Activations