INDEX

Explanations

terms related to reckoning and accountability

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

 escorte

-0.17

ÑĢÐ³

-0.16

elow

-0.16

_reaction

-0.16

ished

-0.16

omatic

-0.15

rane

-0.15

.Generated

-0.15

COPY

-0.15

ANTE

-0.15

POSITIVE LOGITS

oning

0.34

oned

0.26

lessness

0.25

lessly

0.25

less

0.22

ognition

0.20

 reck

0.20

Ð¾Ð¼ÐµÐ½

0.19

LESS

0.17

ons

0.17

Activations Density 0.007%

terms related to reckoning and accountability

No Comments

No Known Activations