INDEX

Explanations

words related to self-reference and identity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

ãĥŃãĥ¼

-0.18

ãĥĥãĤ«ãĥ¼

-0.15

 Rough

-0.15

atak

-0.14

abet

-0.14

chalk

-0.14

RYPTO

-0.14

wat

-0.14

secutive

-0.13

 Blank

-0.13

POSITIVE LOGITS

agrid

0.17

opc

0.17

assin

0.16

 Kens

0.16

iversal

0.15

chaft

0.15

gett

0.15

ALCHEMY

0.15

pta

0.15

essed

0.14

Activations Density 0.037%

words related to self-reference and identity

No Comments

No Known Activations