INDEX

Explanations

occurrences of specific names or identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

loud

-0.15

presence

-0.15

rego

-0.14

rech

-0.14

overrides

-0.14

LOUD

-0.14

rowable

-0.14

viewer

-0.14

Toe

-0.14

ikan

-0.13

POSITIVE LOGITS

xious

0.23

things

0.20

veau

0.20

urnal

0.19

ises

0.18

thern

0.17

AlgorithmException

0.17

theast

0.17

elia

0.16

ce

0.16

Activations Density 0.032%

occurrences of specific names or identifiers

No Comments

No Known Activations