INDEX

Explanations

references to the word "back" in 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 @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

LOCKS

-0.15

abelle

-0.15

illard

-0.15

xbb

-0.14

abel

-0.14

 frau

-0.14

eka

-0.14

-vous

-0.14

raison

-0.14

opot

-0.14

POSITIVE LOGITS

slash

0.19

side

0.18

ness

0.18

/back

0.17

ronym

0.17

iw

0.17

slashes

0.16

eo

0.16

ed

0.16

NOWLED

0.16

Activations Density 0.068%

references to the word "back" in various contexts

No Comments

No Known Activations