INDEX

Explanations

specific mentions of records or recordings

oai_token-act-pair · gpt-3.5-turbo

terms related to recovery or recuperation processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 flawed

-0.65

ahime

-0.62

MENTS

-0.61

Dh

-0.58

 bordering

-0.58

 Cruiser

-0.58

 smoke

-0.57

 Nicarag

-0.57

 sting

-0.57

 slashing

-0.54

POSITIVE LOGITS

ipient

1.37

ords

1.36

ruit

1.30

ursive

1.27

ycle

1.27

urrent

1.26

onnaissance

1.25

ruits

1.24

ording

1.22

ipes

1.21

Activations Density 0.011%

specific mentions of records or recordings

terms related to recovery or recuperation processes

No Comments

No Known Activations